collective(action(clauses ... - revista jurídica...
TRANSCRIPT
1195
COLLECTIVE ACTION CLAUSES FOR PUERTO RICAN BONDS: BORROWING COSTS, PRACTICAL CONSIDERATIONS AND
LESSONS FROM SOVEREIGN DEBT
ARTICLE
ANTONIO J. PIETRANTONI*
Introduction ............................................................................................................ 1195 I. Background Information .................................................................................... 1201
A. C.A.C.s as Living, Breathing Contract Terms ............................................ 1202 B. The History of C.A.C.s ................................................................................ 1205 C. C.A.C.s in the Sovereign and Municipal Debt Contexts ......................... 1208
i. The Good, the Bad and the Ugly ......................................................... 1208 ii. The Implications of Sovereign Immunity ............................................ 1211 iii. Federal and State Law Considerations .............................................. 1214
II. Survey of Studies on Borrowing Costs from Using C.A.C.s ............................ 1216 A. Early Studies: Eichengreen & Mody (2000 and 2004), Gugiatti &
Richards (2003), Becker, Richards & Thaicharoen (2003) ....................... 1216 B. Recent Studies: Choi, Gulati & Posner (2011), Bardozzetti &
Dottori (2013), Bradley & Gulati (2013) ..................................................... 1220 IV. The C.A.C. in Practice and Recent Developments ......................................... 1224
A. The Belizean Experience ............................................................................. 1224 B. Recent Developments and Proposals for Reform ..................................... 1229
V. C.A.C.s in Puerto Rican Bonds: Implications for the Future ........................ 1232
INTRODUCTION
* J.D. candidate, 2015, New York University School of Law; B.A. cum laude in History, 2010, Cornell University. This article reflects a modified version of work submitted for Professor Ana De-‐mel’s Financing Development Seminar during the fall semester of 2014. I would like to thank Profes-‐sor Demel for the opportunity to research and write about a topic that is particularly relevant to the economic reality of my homeland, the Puerto Rico my generation will inherit. I would also like to thank Clayton Gillette, Max E. Greenberg Professor of Contract Law at New York University School of Law, for his invaluable guidance on related topics, and my family for their constant support, in par-‐ticular my father, Roberto Pietrantoni, for his thoughtful comments and feedback. Lastly, I’m im-‐mensely grateful for the comments and edits I received from my friends and classmates Hillary Coleman and Fernando Martínez. This article is product of academic work as a student, opinions and errors are my own.
1196 REVISTA JURÍDICA UPR Vol. 84
EBT CAN BRING ANY COUNTRY TO ITS KNEES. IN A SHORT AMOUNT OF TIME, economic fortunes can turn for the worse. A country can lose control of its financial equilibrium and its social, political and economic
stability can quickly turn dour. Nevertheless, the financial obligations incurred in the past will remain, relentlessly knocking on the door as grudging reminders of promises made during better days and different circumstances. The quaint illusion that twist and shout spending and Mickey Mouse economics could last in perpetuity will be abruptly shattered. Reality hits when deficits mount, the market’s goodwill evaporates and governments teeter on the verge of a precipice. These problems complicate a nation’s ability to honor its outstanding debt obli-‐gations, critical to financing day-‐to-‐day expenditures and providing essential public services, while citizens bear the brunt of the fallout from the remedial measures conjured up to stem the tide of distress. For more than a generation, one could have been fooled into thinking that these problems distinctively be-‐longed in the shores of the Aegean, the grasslands of the Pampas, or some hap-‐less, yet discrete, locality like Detroit. The truth is that American states are not completely immune from this reality, regardless of their status as quasi-‐sovereign entities within a federal system.1 Neither is the Commonwealth of Puerto Rico.2
States3 and sovereign nations are subject to similar economic, social and po-‐litical constraints.4 Like sovereigns, states are tasked with protecting the public welfare of their citizens and borrowing money from the public debt markets to finance expenditures. To be sure, sovereigns are different from states in several critical aspects,5 the most conspicuous being that sovereigns wield substantial control over their own legal systems while states are subject to the overriding power of the Federal Constitution and federal laws. However, governance re-‐sponsibilities, sources of revenue and the underlying market forces governing their debt are essentially the same. Both sovereigns and states are awkwardly constrained by the fact that their revenue raising capacity is limited by political and social concerns. No politician wants their historical legacy to elicit unpleas-‐ant memories of higher taxes or painful austerity measures. In addition, there is
1 Throughout this article, American state, American municipal and American jurisdiction refers to the states, municipalities and jurisdictions, respectively, of the United States of America.
2 Throughout this article, Puerto Rico and the Commonwealth will be used interchangeably in reference to the political entity known as the Commonwealth of Puerto Rico. While Puerto Rico is a not a state, it is an unincorporated federal territory subject to the plenary powers of Congress. How-‐ever, I will ignore Puerto Rico’s current status and treat it like a state because Puerto Rico’s political status is irrelevant to the arguments I put forward in this article.
3 Throughout this article, states refers to the states of the United States of America.
4 See Adam Feibelman, American States and Sovereign Debt Restructuring, in WHEN STATES GO BROKE 146 (Peter Conti-‐Brown & David A. Skeel, Jr., eds., 2012) (arguing that states can learn from the experience of sovereign nations because they share structural similarities).
5 For example, American states do not need to internalize the losses from firms within their jurisdictions to the same extent as sovereigns (i.e., bank exposures) and sovereigns are subject to market pressures on monetary affairs such as floating exchange rate fluctuations.
D
Núm. 4 (2015) COLLECTIVE ACTION CLAUSES FOR PUERTO RICAN BONDS 1197
a point where higher marginal tax rates will lead to inefficient results, spurring an exodus followed by a diminishing rate of tax revenue due to a shrinking tax base.6 Be that as it may, states have little modern debt restructuring experience.7 The experience of sovereign nations therefore furnishes useful insight of what states can expect when facing debt crises.
The Commonwealth of Puerto Rico’s current juncture is not too distant from the scenario discussed above. Since 1952, Puerto Rico has been a Commonwealth that shares common defense, market and currency with the United States.8 The Commonwealth’s debt obligations have traditionally been a staple of any munic-‐ipal investor’s portfolio,9 given its triple-‐tax exempt status from federal, state and local taxes.10 Since 2006, the Commonwealth has endured a sustained economic recession leading to a considerable population decline and economic contrac-‐
6 See, e.g., ARTHUR LAFFER, THE LAFFER CURVE: PAST, PRESENT, AND FUTURE (2004), available at www.heritage.org/research/reports/2004/06/the-‐laffer-‐curve-‐past-‐present-‐and-‐future (describing the relatively undisputed canon of tax policy of how tax rates impact taxpayer behavior).
7 A number of states, however, went through a series of debt crises and repudiations during the nineteenth century. See BENJAMIN ULYSSES RATCHFORD, AMERICAN STATE DEBTS (1941), available at http://babel.hathitrust.org/cgi/pt?id=mdp.39015021559052;view=1up;seq=22 (discussing the experi-‐ence of multiple American states with debt crises during the nineteenth century and their different approaches to solving them); WILLIAM A. SCOTT, THE REPUDIATION OF STATE DEBTS: A STUDY IN THE FINANCIAL HISTORY OF MISSISSIPPI, FLORIDA, ALABAMA, NORTH CAROLINA, SOUTH CAROLINA, GEORGIA, LOUISIANA, ARKANSAS, TENNESSEE, MINNESOTA, MICHIGAN, AND VIRGINIA (1893) (presenting an over-‐view of the constitutional, legal, financial and philosophical aspects of the state debt repudiations that occurred throughout the nineteenth century). Arkansas, during the 1930’s, is the only known modern example of a state engaging creditors to renegotiate its obligations. See Monica Davey, The State That Went Bust, N.Y. TIMES (Jan. 22, 2011), http://www.nytimes.com/2011/01/23/weekinreview/ 23davey.html.
8 Puerto Ricans are citizens of the United States, but do not vote in national elections and do not pay most federal taxes.
9 The Commonwealth issues general obligation bonds backed by its full faith and credit. The Commonwealth’s instrumentalities, such as the Puerto Rico Electric Power Authority (P.R.E.P.A.) and the Puerto Rico Aqueduct and Sewer Authority (P.R.A.S.A.), issue debt in the form of revenue bonds payable from their revenues. See Tax-‐Exempt Securities by Issuer, GOV’T DEV. BANK FOR P.R., http://www.gdb-‐pur.com/investors_resources/exempt-‐securities.html (last visited Apr. 12, 2015). For the sake of convenience, I will treat the Commonwealth, and its public instrumentality issuers, as if they were the same entity, unless otherwise noted. Revenue and general obligation debt differ from one another in multiple ways that can affect their restructuring options, but these differences are not all that important for the purposes of this article.
10 The Puerto Rico Federal Relations Act provides that:
All bonds issued by the Government of Puerto Rico, or by its authority, shall be exempt from taxation by the Government of the United States, or by the Government of Puerto Ri-‐co or of any political or municipal subdivision thereof, or by any State, Territory, or pos-‐session, or by any county, municipality, or other municipal subdivision of any State, Terri-‐tory, or possession of the United States, or by the District of Columbia.
Puerto Rico Federal Relations Act, 48 U.S.C. § 745 (2012).
1198 REVISTA JURÍDICA UPR Vol. 84
tion.11 A salient feature of this recession has been the high debt burdens Puerto Rico and its instrumentalities carry and, in particular, a public debate on renego-‐tiating these obligations in order to restore them to sustainable levels.12 The cata-‐lyst for this debate occurred in February of 2014, when the three major credit rating agencies downgraded the credit of every single Commonwealth issuer to below investment grade, colloquially known as junk status.13 Since then, the Is-‐land’s access to the capital markets has substantially narrowed. Commentators have compared Puerto Rico’s predicament to the likes of Greece and Argentina.14
These comparisons paint a daunting picture. Argentina’s default in 2002, and the never-‐ending litigation it faces from holdout creditors demanding full repayment, continues to capture headlines and simmer emotions.15 A mere handful of years ago, an over-‐indebted Greece threatened to bring down Eu-‐rope’s monetary union experiment. Notwithstanding a historic debt restructur-‐ing in 2012, concerns over Greece’s debt burdens, fiscal challenges and its poten-‐tial exit from the monetary union abound to this day.16 Throughout these epi-‐sodes, the lack of a clear legal or contractual framework that could be relied up-‐
11 See generally FED. RESERVE BANK OF N.Y., AN UPDATE ON THE COMPETITIVENESS OF PUERTO RICO’S ECONOMY (2014), available at http://newyorkfed.org/outreach-‐and-‐education/puerto-‐rico/2014/Puerto-‐Rico-‐Report-‐2014.pdf.
12 E.g., Editorial, Reestructuración, no hay otra salida, EL NUEVO DÍA (Feb. 3, 2015), http://www.elnuevodia.com/opinion/editorial/nota/reestructuracionnohayotrasalida-‐2002795/ (arguing that debt restructuring is necessary for Puerto Rico’s economic recovery).
13 Global Credit Research, Moody’s Downgrades Puerto Rico GO and Related Bonds to Ba2, Notched Bonds to Ba3 and COFINA Bonds to Baa1, Baa2; Outlook Negative, MOODY’S (Feb. 7, 2014), https://www.moodys.com/research/Moodys-‐downgrades-‐Puerto-‐Rico-‐GO-‐and-‐related-‐bonds-‐to-‐Ba2-‐-‐PR_292399; STANDARD & POOR’S, PUERTO RICO GO RATING LOWERED TO ‘BB+’; REMAINS ON WATCH NEGATIVE (2014), available at www.standardandpoors.com/ratingsdirect; Fitch Downgrades Puerto Rico GO and Related Debt Ratings to ‘BB’; Outlook Negative, REUTERS (Feb. 11, 2014), http://www.reuters.com/article/2014/02/11/ny-‐fitch-‐ratings-‐pr-‐idUSnBw116233a+100+BSW20140211.
14 See, e.g., Greece in the Caribbean, ECONOMIST (Oct. 26, 2013), http://www.economist.com/ news/leaders/21588374-‐stuck-‐real-‐debt-‐crisis-‐its-‐back-‐yard-‐america-‐can-‐learn-‐europes-‐aegean; An-‐drew Bary, Troubling Winds, BARRON’S (Aug. 26, 2013), http://online.barrons.com/articles/ SB50001424052748704719204579022892632785548; Michael Connor & Tom Hals, Despite Blockbuster Bond Sale, Puerto Rico Debt-‐holders Still in Crosshairs, REUTERS (Mar. 14, 2014), http://www.reuters.com/article/2014/03/14/usa-‐puertorico-‐restructuring-‐idUSL2N0MA29D20140314.
15 Argentina imposed haircuts between seventy one percent to seventy five percent. Feibelman, supra note 4, at 166 (discussing Argentina). To this day, the country has not been able to reach an agreement with holdout creditors and the final resolution of the litigation is still pending. See J. F. HORNBECK, CONG. RESEARCH SERV., R41029, ARGENTINA’S DEFAULTED SOVEREIGN DEBT: DEALING WITH THE “HOLDOUTS” 2 (2013), available at https://www.fas.org/sgp/crs/row/R41029.pdf; Peter Eavis & Alexandra Stevenson, Argentina Finds Relentless Foe in Paul Singer’s Hedge Fund, N.Y. TIMES (July 30, 2014, 10:35 PM), http://dealbook.nytimes.com/2014/07/30/in-‐hedge-‐fund-‐argentina-‐finds-‐relentless-‐foe/.
16 See, e.g., Landon Thomas Jr. & Jack Ewing, Greek Debt Standoff Awaits a Decisive Move, N.Y. TIMES (Feb. 12, 2015, 5:47 PM), http://dealbook.nytimes.com/2015/02/12/greek-‐debt-‐standoff-‐awaits-‐a-‐decisive-‐move/; Landon Thomas Jr., Greece Flashes Warning Signals About its Debt, N.Y. TIMES (Apr. 19, 2015), http://www.nytimes.com/2015/04/20/business/international/greece-‐flashes-‐warning-‐signals-‐about-‐its-‐debt.html.
Núm. 4 (2015) COLLECTIVE ACTION CLAUSES FOR PUERTO RICAN BONDS 1199
on to regain sustainable levels of debt erected an immense obstacle. Many pit-‐falls could have been avoided had these frameworks been in place.
Unlike individuals and businesses, sovereign nations and American states cannot file for bankruptcy protection.17 To be sure, a limited portion of the American municipal debtors can secure relief under chapter 9 of the Bankruptcy Code, the traditional mechanism available for certain municipal issuers.18 As of the time of this writing, however, Puerto Rican municipalities and public corpo-‐rations are not eligible to file for chapter 9 protection, although a congressional bill has been introduced to correct this unexplained exclusion.19 Regardless, chapter 9 is not available to states, whose definition includes Puerto Rico,20 be-‐cause states are excluded from the definition of municipalities.21 Sovereign na-‐tions, American states and every Commonwealth issuer must instead rely on ad hoc debt restructurings outside the protection of a formal bankruptcy mecha-‐nism. As a result, the success of a restructuring depends on carefully navigating the treacherous debt markets and reaching a solution calibrated to address the interests of both creditors and debtors.
The desire to minimize the economic and reputational costs incurred by re-‐structuring leads many to believe that sovereigns would actively attempt to ne-‐gotiate with creditors a remedy amenable to both.22 Doing so allows the issuer to cabin any negative externalities associated with a restructuring and pave the way for a speedy reentry into the debt markets. However, an isolated group of credi-‐tors can hold out and actively challenge these attempts. Moreover, bondholders may resist because of fears that the sovereign debtor is behaving opportunistical-‐ly, shifting losses onto them without absorbing its share of pain. In any case, the critical obstacle is the need to secure creditor coordination and bind a minority of dissenters. Put simply, they need to overcome a collective action problem.
This article will focus on collective action clauses (C.A.C.s) and their role as a contractual solution to the bondholder coordination problem. Foregoing una-‐nimity, these clauses permit changes to certain bond terms after a majority of bondholders assents to the modification. In particular, C.A.C.s allow changes to
17 There is no international sovereign bankruptcy mechanism and chapter 9 of the Bankrupt-‐cy Code is not available to states because they are not eligible municipalities. Bankruptcy Code, 11 U.S.C. § 109(c)(1) (2012) (“An entity may be a debtor under chapter 9 of this title if and only if such entity . . . is a municipality . . . .”). Id. § 101(40) (“The term municipality means political subdi-‐vision or public agency or instrumentality of a State.”).
18 Id. §§ 901-‐946.
19 Puerto Rico Chapter 9 Uniformity Act of 2015, H.R. 870, 114th Cong. (2015).
20 11 U.S.C. § 101(52). “The term State includes the District of Columbia and Puerto Rico, except for the purpose of defining who may be a debtor under chapter 9 of this title.” Id.
21 Id. § 101(40).
22 FEDERICO STURZENEGGER & JEROMIN ZETTELMEYER, DEBT DEFAULTS AND LESSONS FROM A DECADE OF CRISES 35 (2006) (discussing the impact of a default on the borrower’s reputation); Richard M. Hynes, State Default and Synthetic Bankruptcy, 87 WASH. L. REV. 657, 673 (2012) (discussing how states value a reputation for honoring commitments.).
1200 REVISTA JURÍDICA UPR Vol. 84
the payment terms of outstanding bonds —such as interest, maturity and prin-‐cipal amount— under the guise of a democratic framework that incentivizes debtors to engage creditors in a diplomatic manner.23 Absent C.A.C.s and formal bankruptcy mechanisms, debt relief negotiations can be messy and conten-‐tious.24 C.A.C.s, therefore, provide the functional equivalent of an orderly negoti-‐ation forum where Contract Law and market forces fill the void left behind by bankruptcy’s absence.25
In contrast to sovereign bonds, C.A.C.s are almost universally absent in bonds issued by U.S. states and their instrumentalities.26 Puerto Rican bonds are no exception. A cursory examination of terms included in several outstanding Puerto Rican general obligation and public corporation revenue bond issuances reveals that the current contractual framework is of limited utility to address the collective action problem. These bonds utterly lack a contractual mechanism that would permit the Commonwealth to renegotiate payment terms without the need to secure the consent of every single bondholder.27 The Federal Gov-‐ernment’s past stance on this matter evinces some acknowledgement of the val-‐ue of C.A.C.s. Faced with a similar dilemma in the sovereign debt context at the turn of the twenty-‐first century, the United States Department of the Treasury supported C.A.C.s as a way to provide future restructurings with a market-‐oriented solution.28
In the absence of a formal bankruptcy alternative, Puerto Rican policymak-‐ers should consider including C.A.C.s in all future Commonwealth bond issuanc-‐
23 E.g., LEE C. BUCHHEIT, HOW TO NEGOTIATE EUROCURRENCY LOAN AGREEMENTS 145-‐49 (2d ed. 2006) (describing how majority amendment clauses work in practice).
24 David A. Skeel Jr., States of Bankruptcy, 79 U. CHI. L. REV. 677, 726 (2012).
25 The assumption I make is one that reflects the current state of affairs. Even if Puerto Rican municipalities were extended chapter 9 eligibility, presumably removing their need for C.A.C.s, the Commonwealth’s general obligations would still be able to benefit from their inclusion, as discussed in infra Part I.C.iii. This latter point on the general obligation bonds hinges on the fact that there is no restructuring alternative for debts issued directly by states.
26 See Steven L. Schwarcz, A Minimalist Approach to State “Bankruptcy”, 59 UCLA L. REV. 322, 329-‐31 (2011-‐2012) (noting the absence of C.A.C.s in state bonds).
27 A general examination of certain general obligation and revenue bond official statements reveals that majority amendment provisions are essentially non-‐existent. If anything, the existence of accelera-‐tion clauses makes the framework all the more schizoid. See $673,145,000, PUERTO RICO ELECTRIC POWER AUTHORITY, POWER REVENUE BONDS, SERIES 2013A (2013), available at http://www.gdb-‐pur.com/investors_resources/documents/2013-‐08-‐19-‐OSSeries2013-‐A-‐Agosto2013-‐FINAL.pdf; $2,318,190,000, COMMONWEALTH OF PUERTO RICO, PUBLIC IMPROVEMENT REFUNDING BONDS, SERIES 2012 A, (GENERAL OBLIGATION BONDS) (2012), available at http://www.gdb-‐pur.com/investors_resources/documents/PRCommonwealth01a-‐FIN.pdf; $1,800,450,000, PUERTO RICO AQUEDUCT AND SEWER AUTHORITY, REVENUE BONDS, SERIES 2012A (SENIOR LIEN) (2012), available at http://www.gdb-‐pur.com/investors_resources/documents/PRAqueductSewerAuth01a-‐FIN-‐1800MM.pdf.
28 See Sean Hagan, Designing A Legal Framework to Restructure Sovereign Debt, 36 GEO. J. INT’L L. 299, 390-‐94 (2004-‐2005) (discussing the United States Department of the Treasury’s preference for C.A.C.s and its opposition to a standardized restructuring mechanism proposed by the International Monetary Fund).
Núm. 4 (2015) COLLECTIVE ACTION CLAUSES FOR PUERTO RICAN BONDS 1201
es. To be sure, these clauses will not obliterate the collective action problems that fuels holdout creditors, but they help equalize the playing field when properly designed. Notwithstanding the possibility of increased borrowing costs in the short-‐term, C.A.C.s can minimize collective action problems and make any cost incurred by their inclusion worth the hassle down the road. These clauses ease and streamline debt workouts by preventing the grueling trench warfare between the Commonwealth and its bondholders that would otherwise occur if negotiated debt relief hinged on the unanimous consent of every affect-‐ed bondholder. Future stakeholders will benefit from a clear and pragmatic con-‐tractual mechanism successfully tested in the sovereign debt context.
The legal issues surrounding the current debt crisis, and the remedies em-‐ployed by the Commonwealth to address it, are beyond the scope of this article. Instead, I will employ a prospective, forward-‐looking approach, and discuss the implications of including C.A.C.s in Puerto Rico’s future issuances. In particular, I will: (1) provide a brief background of C.A.C.s within the sovereign and United States municipal debt context; (2) survey empirical research on the borrowing costs C.A.C.s may impose; (3) discuss recent developments, including their use by Belize —the only modern New York Law sovereign issuer to activate C.A.C.s as part of a debt restructuring process— and proposals advanced by the Interna-‐tional Monetary Fund (I.M.F.) pushing for the strengthening of existing C.A.C.s, and (4) analyze C.A.C.s as viable tools for the Commonwealth’s future debt man-‐agement.
I . BACKGROUND INFORMATION
As a threshold matter, the contractual framework of sovereign debt reflects the reality that there is an inherent element of risk in loaning money to a sover-‐eign nation that is subject to political pressures,29 and likewise responds to a variety of constituent interests.30 Notwithstanding differences in politics and economics, the analogous framework of state debt reflects similar considera-‐tions.31 The content of bond contracts plays an important role when a dispute
29 See Stephen J. Choi et al., Political Risk and Sovereign Debt Contracts (Univ. of Chi. Law Sch., John. M. Olin Law & Econ., Working Paper No. 583, Pub. Law & Legal Theory, Working Paper No. 370, 2011), available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1962788 (discussing how sovereign bond contracts are inherently risky because they are subject to the incomplete contract problem whereby they will not always be able to respond to shifting states of the world and the polit-‐ical responses to them).
30 See Stephen J. Choi et al., Pricing Terms in Sovereign Debt Contracts: A Greek Case Study with Implications for the European Crisis Resolution Mechanism, 6 CAPITAL MKTS. L.J. 163, 169-‐70 (2011) [hereinafter Choi et al., Pricing Terms] (discussing the various social, political and economic pressures sovereign debtors are subject to).
31 In their book, Robert S. Amdursky, Clayton P. Gillette and G. Allen Bass discuss this tension in the municipal debt context:
The governmental issuer, however, presents the investor with substantial risks by virtue of its political position. . . . [C]hanges in circumstances may affect the . . . willingness or abil-‐
1202 REVISTA JURÍDICA UPR Vol. 84
between a creditor and debtor arises. For example, this framework outlines the legal rights and remedies of both parties in case of breach.32 While securing a legal remedy against a sovereign nation has practical limitations, and American states and Puerto Rico may enjoy sovereign immunity protection,33 courts have not shied away from interpreting and enforcing debt contract provisions at the behest of a bondholder.34
State and sovereign debt contracts are also uniquely imbued with an implicit component not readily visible to the naked eye. As opposed to corporate debt, where an entity contracts with specific creditors, state and sovereign debt con-‐tracts have an additional stakeholder: the citizens of the debtor. While citizens are not explicit signatories to the bond agreement, their government is. A debt contract therefore allocates to these third parties the benefits of borrowing and the burdens of distress by virtue of their government’s pawning sources of reve-‐nue every time it borrows. Because the purpose of government is to provide for the health, safety and welfare of its citizens, an inability to access capital can be disastrous from a net welfare and economic perspective. A government is, there-‐fore, the prime example of something that is too big to fail.
A. C.A.C.s as Living, Breathing Contract Terms
C.A.C.s have gained mass appeal in the world of sovereign finance in the last few years. They previously had a taciturn and less glamorous existence as a bor-‐ing, mechanical clause rarely included in New York Law sovereign bonds.35 In a world where financial fortunes can rapidly change, and the pool of sovereign creditors is heterogeneous and widely dispersed, C.A.C.s have played a leading role as a practical, contract-‐oriented response to the realities of the modern se-‐curities market. In particular, C.A.C.s respond to the fact that sovereign bonds are liquid and tradable multi-‐creditor instruments where the pool of bondhold-‐
ity of the municipal issuer to pay principal and interest. As a public body, the issuer may consider that its primary obligation runs to its constituents, just as a private corporation’s primary obligation is to its stockholders.
ROBERT S. AMDURSKY ET AL., MUNICIPAL DEBT FINANCE LAW, THEORY AND PRACTICE 317 (2d ed. 2013).
32 See W. Mark C. Weidemaier & Mitu Gulati, Sovereign Debt and the “Contracts Matter” Hypoth-‐esis, in OXFORD HANDBOOK OF LAW AND ECONOMICS (forthcoming 2014) [hereinafter Weidemaier & Gulati, Sovereign Debt] (manuscript at 4), available at http://scholarship.law.duke.edu/ faculty_scholarship/3380/ (discussing the importance of bond terms).
33 See discussion infra Part I.C.ii.
34 See Weidemaier & Gulati, Sovereign Debt, supra note 32 (discussing the enforcement of bond-‐holder rights in the sovereign debt context). For a discussion on how municipal bondholder rights are enforced by United States courts while grappling with the difficult legal and policy questions these rights elicit, see generally AMDURSKY ET AL., supra note 31, at 333-‐90, 413-‐14 (providing background on the multiple ways United States courts have interpreted and enforced municipal bondholder rights).
35 David Billington, European Collective Action Clauses, in SOVEREIGN DEBT MANAGEMENT 399 (Rosa M. Lastra & Lee Buchheit eds., 2014).
Núm. 4 (2015) COLLECTIVE ACTION CLAUSES FOR PUERTO RICAN BONDS 1203
ers has fluctuating identities and different objectives.36 As a result, C.A.C.s are part of a concerted effort to check the “grab and run instinct of each bondholder . . . long enough to permit a coordinated workout,”37 and prevent the “tyranny of the minority” to usurp the collective will of a majority that generously agreed, likely because it is in their best interest, to grant debt relief.38
Although they have multiple incarnations, C.A.C.s overwhelmingly incorpo-‐rate the same characteristics. Generally speaking, by buying a bond that includes a C.A.C., an investor is implicitly agreeing that his “legal and economic rights . . . can be altered without his consent if” enough fellow bondholders agree to the alteration and that his individual enforcement power will be weaker if a majority of fellow bondholders “restrain him.”39 Bondholders may understandably have qualms with the above, but they should keep in mind that the costs also come with benefits: while bondholders can rightly worry that they might have to ac-‐cept modifications they do not agree to, they should also worry that other bond-‐holders could obstruct a modification that they would willingly accept.40
Therefore, C.A.C.s require a specific consent threshold (i.e., voting majority) in order to amend the terms of outstanding bonds. Thresholds tend to be based on the aggregate principal amount outstanding at the time an amendment is proposed, and generally require a quorum of bondholders for specific reserved matters. These matters can be anything from maturity dates, interest rates, amortization schedules and principal amount. Thresholds range from as low as 18.75% to as high as 85%, but the typical requirement tends to be 75%, and the European Union recently adopted a standardized consent threshold requirement of 66.67% for all bonds issued by Eurozone borrowers.41 Without a doubt, the choice of a consent threshold has significant implications: “Set the threshold too high and the usefulness of the CAC is diluted; set it too low and investors may be reluctant to subscribe [to the bond] for fear of the ease with which other bond-‐
36 See Lee C. Buchheit & G. Mitu Gulati, Sovereign Bonds and the Collective Will, 51 EMORY L.J. 1317, 1320-‐23 (2002) [hereinafter Buchheit & Gulati, Sovereign Bonds] (discussing the reasons that led to the success C.A.C.s).
37 Id. at 1321.
38 Id. at 1336.
39 Billington, supra note 35, at 400.
40 In his work, David Billington argues that:
Each bondholder may be worried that harmful modifications could be made to the terms of their bonds against their will. But they should be equally worried that they could get stuck in a situation in which they wish to accept whatever is being offered to them . . . and a small holdout creditor causes deadlock or precipitates a disorderly situation by bringing proceedings against the issuer.
Id.
41 Michael Bradley & Mitu Gulati, Collective Action Clauses for the Eurozone, 18 REV. FIN. (forth-‐coming Oct. 2014) (manuscript at 4), available at http://scholarship.law.duke.edu/ faculty_scholarship/2455/. A subsequent version was published in 2014. See Michael Bradley & Mitu Gulati, Collective Action Clauses for the Eurozone, 18 REV. FIN. 2045 (2014).
1204 REVISTA JURÍDICA UPR Vol. 84
holders could approve changes to the terms of the bonds.”42 Consequently, an optimal collective action threshold may bear directly on an issuer’s cost of capi-‐tal.43 Although a discussion on appropriate C.A.C. voting thresholds merits a separate paper, the current prevalent market practice appears to be a consent threshold of seventy five percent.44
Investors naturally worry that opening the door to easier debt restructuring can lend itself to abuse. To address these concerns, issuers have designed C.A.C.s with protections that go beyond raw45 voting thresholds.46 Some versions specifi-‐cally disenfranchise certain bondholders either because they are public instru-‐mentalities of the issuer or intimately connected to the state apparatus. Other C.A.C.s incorporate an aggregation mechanism where consent thresholds are measured across a set of eligible bonds, as opposed to an issue-‐specific thresh-‐old. The underlying reasoning behind aggregation is that bondholders will be reluctant to accept a haircut without “any guarantee that the holders in the other classes will provide similar relief,” and that “it is fair for holders of a single series of bonds to be bound by a proposed modification if the modification is approved by a supermajority of all other affected investors.”47 To quell concerns about ag-‐gregation’s potential for abuse, some C.A.C.s have independent quorum or con-‐sent requirements within each individual issuance. Without a doubt, these varia-‐tions bear upon how issuers and bondholders perceive the C.A.C. Subtle differ-‐ences can shift substantial control over the restructuring process to either side, and the particular nuances of the outstanding debt stock of each Commonwealth issuer should be considered when structuring these clauses that “have evolved from a short and simple one-‐paragraph clause . . . to a complex multi-‐page ani-‐mal . . . .”48
42 Billington, supra note 35, at 406 (discussing the Goldilocks dilemma).
43 See generally Jenna Seki, Optimal Collective Action Clause Thresholds (May 14, 2007) (un-‐published thesis, Stanford University), available at https://economics.stanford.edu/ files/Theses/Theses_2007/Seki2007.pdf; Andrew G. Haldane et al., Optimal Collective Action Clause Thresholds (Bank of Eng., Working Paper No. 249, 2004), available at http://www.bankofengland.co.uk/research/Documents/workingpapers/2005/wp249.pdf.
44 Stephen J. Choi et al., The Dynamics of Contract Evolution, 88 N.Y.U. L. REV. 1, 12 (2013) [here-‐inafter Choi et al., Contract Evolution]. However, the specific identity of the bondholder pool should inform this decision. As a matter of policy, having a concentrated pool of bondholders may merit having a lower vote threshold since it would be easier for bondholders to acquire a blocking position. Likewise, having a heterogeneous and dispersed bondholder pool may make it hard for the Com-‐monwealth to contact enough bondholders to approve a restructuring via the C.A.C.s. See discussion infra Part IV.
45 Choi et al., Contract Evolution, supra note 44, at 13.
46 A recent study identified ten “CAC models . . . used during the 1990-‐2011 period.” Id. at 18 (dis-‐cussing the dimensions in which CACs tend to diverge).
47 David Billington, supra note 35, at 409.
48 Choi et al., Contract Evolution, supra note 44, at 13.
Núm. 4 (2015) COLLECTIVE ACTION CLAUSES FOR PUERTO RICAN BONDS 1205
B. The History of C.A.C.s
Modern C.A.C.s originated in the English debt markets of the nineteenth century, amidst the rapid expansion of commerce ignited by the Industrial Revo-‐lution. This coincided with a better understanding of how multi-‐creditor debt instruments, and their network of heterogeneous investors with idiosyncratic interests and motivations, posed a set of distinct problems when an issuer’s fi-‐nances turned precarious. Bond issuers and investors “c[a]me to believe that bondholder cooperation in a distressed situation was highly desirable.”49 In those days, bankruptcy was a rather inflexible mechanism that meant liquidation and recovery of only a portion of an investment. There was no legal framework through which bondholders could collectively agree to “temporary deferment of their claims or a partial reduction in the amounts due” that would have “pre-‐serve[d] their debtor as a going concern from whom payments, even if late or . . . [modified], could be expected.”50 A tool was needed to bind every bondholder, avoid the liquidation of the debtor and secure some continued stream of pay-‐ments. It was unfair for a group of stiff-‐necked creditors to receive full payment of their claims partly on account of another group of generous creditors who were understanding of the debtor’s plight.51 A majority action clause was created to solve this quandary: a distressed debtor could modify payment terms and bind minority dissenters after a supermajority of those affected agreed.52 The clause became wildly popular in the London debt markets soon after its inclusion with-‐in a book of corporate forms in the late 1870’s.53 They have been a staple of sov-‐ereign bonds issued under English Law ever since.
The United States securities markets did not readily embrace the English so-‐lution. It was not until after Mexico included a C.A.C. in a bond issued in the spring of 2003 that sovereign bonds under New York Law began to embrace them. While multiple motivations have been advanced to explain Mexico’s deci-‐sion to adopt these clauses,54 one thing is certain: given the market’s tenuous relationship with C.A.C.s at the time, Mexico’s large influence in the emerging debt markets and solid credit rating fit the bill for an experiment in revolutioniz-‐ing boilerplate bond terms.55 Prior to Mexico’s trailblazing move, the general assumption was that sovereigns could not alter the payment terms of outstand-‐
49 See Buchheit & Gulati, Sovereign Bonds, supra note 36, at 1321.
50 Id. at 1324.
51 Id.
52 Id. at 1324-‐25.
53 Id. (discussing the origins of the English majority action clauses).
54 See Anna Gelpern & Mitu Gulati, Public Symbol in Private Contract: A Case Study, 84 WASH. U. L. REV. 1627 (2006) (discussing the multiple official and unofficial explanations that can explain the C.A.C.’s rise to prominence).
55 Id. at 1678 (explaining why Mexico was able to motivate other sovereigns to incorporate C.A.C.s in subsequent debt issuances).
1206 REVISTA JURÍDICA UPR Vol. 84
ing bonds without the consent of every bondholder. According to scholars, the American stubbornness to include these clauses did not reflect doubts about their validity; market participants instead “worr[ied] that a provision permitting a post-‐issuance change to payment terms might impair . . . [the] bond’s status as a negotiable instrument under the Negotiable Instruments Law . . . .”56 Conse-‐quently, listing exchanges, issuers and underwriters discouraged their inclusion to avoid a “cloud over their status.”57
The Trust Indenture Act’s prohibition on majority voting clauses with re-‐spect to the payment terms of corporate bonds seems to have also influenced the drafting of sovereign bonds terms to some degree.58 The salient premise behind this prohibition was that bankruptcy is the adequate and preferable solution to fixing distressed corporate finances.59 Given the longstanding existence of a mar-‐ket for corporate bonds, some scholars argue that it is possible that sovereign bonds evolved to mirror corporate bonds unintentionally. Prior to the wide-‐spread inclusion of C.A.C.s, the consent of a majority threshold of bondholders could be obtained to modify certain terms, but unanimous consent was required to alter payment terms.60 Multiple explanations have been offered to explain how this came to be.61 Drafting conventions may have thus been drawn to reflect only terms that United States investors were familiar with and failed to consider cred-‐itor coordination challenges down the road. Likewise, it may have been a mis-‐placed response to the Federal Government’s preference to apply Bankruptcy Law’s priority structure to the corporate context. Under Bankruptcy Law, debt is paid before equity holders obtain any residual value of a corporation’s assets. Nevertheless, the latter reasoning is not applicable to sovereigns and states, where there is no equity, only a finite spectrum of assets that can be liquidated exists and, again, there is no applicable Bankruptcy Law.
Some scholars dispute this history and posit that these market failure narra-‐tives serve those advocating official sector intervention in reforming sovereign lending. They argue, for instance, that prior to C.A.C.s being brought to the fore-‐front of reform debates, creditor coordination problems were tackled via other clauses. For example, some scholars argue that syndicated bank loans had provi-‐sions for modifying their terms and included sharing clauses that allocated losses
56 See Buchheit & Gulati, Sovereign Bonds, supra note 36, at 1326.
57 Id.
58 See Schwarcz, supra note 26, at 330 n.40 (discussing the Trust Indenture Act’s prohibition of C.A.C.s in corporate bonds).
59 W. Mark C. Weidemaier & Mitu Gulati, A People’s History of Collective Action Clauses, 54 VA. J. INT’L L. 51, 59 (2013-‐2014) [hereinafter Weidemaier & Gulati, A People’s History] (discussing how the Trust Indenture Act reflects a preference for Bankruptcy Law as the proper vehicle for a corporate debtor to obtain debt relief).
60 Buchheit & Gulati, Sovereign Bonds, supra note 36, at 1328-‐30 (discussing unanimity require-‐ments).
61 Id. at 1329-‐30 (discussing possibilities for why sovereign bonds required unanimous consent to amend their terms).
Núm. 4 (2015) COLLECTIVE ACTION CLAUSES FOR PUERTO RICAN BONDS 1207
among lenders.62 In addition, collective modification clauses made their way into several sovereign issuances throughout the twentieth century, and a handful of bonds included collective acceleration clauses (i.e., requiring the consent of a certain threshold of bondholders before the bonds could be accelerated).63 These examples, according to this set of scholars, therefore evince some minimal un-‐derstanding of creditor coordination problems prior to the market’s acceptance of C.A.C.s in New York Law sovereign bonds.
Admittedly, it is unclear whether the absence of C.A.C.s in Commonwealth debt issuances is due to flawed assumptions, a historical misunderstanding or a lack of foresight. However, I did not find any reason to believe that the United States municipal debt market has rejected C.A.C.s beyond possible investor re-‐luctance to facilitate any adjustment adverse to their interests. A simple explana-‐tion for their absence in Puerto Rican bonds may be that the earlier investors in Commonwealth debt may have not fathomed a default scenario. Throughout most of the twentieth century, Puerto Rico was a relatively poor agrarian econ-‐omy. Efforts to modernize into a manufacture-‐based economy relied on borrow-‐ing to finance infrastructure projects and economic development. Coupled with artificial stimulus in the form of federal income tax exemptions for United States companies in Puerto Rico,64 it would have been hard to imagine a debt crisis amidst seemingly never-‐ending prospects of economic growth. In a world where attorneys depend on precedents as models for future documentation, reluctance to change bond terms may simply be a symptom of normative complacency.
A study on how sovereign bond contracts have historically evolved identified a series of patterns that may also account for the absence of C.A.C.s in municipal bonds and support the hypotheses discussed in the prior paragraph.65 It takes time for sticky boilerplate contract terms to change; innovations tend to occur only when the debt markets are jolted into realizing that the existing contractual framework may be inadequate.66 Unlike the sovereign debt market, the modern municipal debt market has arguably enjoyed long-‐term stability with little inter-‐ruptions.67 With Puerto Rico, Illinois and California facing fiscal crises, this nor-‐
62 Id. at 1334-‐35 (discussing the prevalence of syndicated bank loans); Weidemaier & Gulati, A People’s History, supra note 59, at 64-‐66 (discussing how collective action problems in the pre-‐C.A.C. era were tackled through sharing clauses in syndicated loans, etc.).
63 Weidemaier & Gulati, A People’s History, supra note 59, at 57-‐58, 65.
64 Up until the completion of its ten-‐year phase out in 2006, I.R.C. § 936 was one of the engines of Puerto Rico’s economy. I.R.C. § 936 (2012). Many Puerto Ricans depended on the manufacturing jobs created by American companies who set up operations in Puerto Rico seeking to benefit from a federal tax code item that allowed them to report tax-‐free income. The expiration of this federal tax item coincided with the beginning of Commonwealth’s current recession.
65 See generally Choi et al., Contract Evolution, supra note 44.
66 Id. at 36.
67 See, e.g., Brian Chappatta, Detroit’s Bankruptcy Doesn’t Faze the Municipal Bond Market, BLOOMBERG BUS. (Aug. 8, 2013), http://www.bloomberg.com/bw/articles/2013-‐08-‐08/detroits-‐bankruptcy-‐doesnt-‐faze-‐the-‐municipal-‐bond-‐market.
1208 REVISTA JURÍDICA UPR Vol. 84
mally temperate market may be due for a series of shocks. It is up to those with a vested interest in resolving them to break away from precedent: “External forces can precipitate a change in the standard, such as the shocks . . . [to] the sover-‐eign bond market . . . . [T]hese shocks lead to an initial period of experimenta-‐tion by more marginal players . . . . Top players have a vested interest in support-‐ing the existing standard through which they maintain their competitive domi-‐nance.”68
C. C.A.C.s in the Sovereign and Municipal Debt Contexts
i. The Good, the Bad and the Ugly
Notwithstanding different narratives on the historical development of C.A.C.s, the debt markets of today pose different challenges than before. Fortu-‐nately, gunboat diplomacy and the use of military might to enforce debt obliga-‐tions is a thing of our fanciful historical past.69 Nevertheless, new challenges arose that could derail economic recovery if left unrestrained.70 The enforcement battles of yore resurfaced in the courtrooms of civilized society. Incorporating C.A.C.s in bond contracts will make it easier for the Commonwealth to deal with these challenges going forward.
As mentioned before, the need to secure creditor coordination and minimize the threat of dissenters stems in part from the dispersed nature of bondholders. It has been further fueled by the rise of specialized distressed debt investors, some of whom are colloquially known as vulture funds.71 These latter funds pur-‐chase distressed sovereign debt at low prices, and subsequently turn them around for a short-‐term profit or, in the extreme, sue to claim their full amount.72 In light of the Commonwealth’s fiscal challenges, distressed debt in-‐vestors have set their eyes on Puerto Rico, supplanting the Commonwealth’s traditional retail bondholder, and amassing significant positions in various out-‐standing debt issuances.73 In addition, media coverage during the recent global
68 Choi et al., Contract Evolution, supra note 44, at 36.
69 Choi et al., Pricing Terms, supra note 30, at 166.
70 Weidemaier & Gulati, Sovereign Debt, supra note 32, at 4 (discussing how the modern strategies bondholders use to pursue legal remedies can unravel a restructuring).
71 E.g., Weidemaier & Gulati, A People’s History, supra note 59, at 83 (describing the rise of specialized holdout creditors known as vulture funds); Samuel E. Goldman, Mavericks in the Mar-‐ket: The Emerging Problem of Hold-‐Outs in Sovereign Debt Restructuring, 5 UCLA J. INT’L L. & FOREIGN AFF. 159 (2000-‐2001) (discussing the holdout creditor problem).
72 Choi et al., Pricing Terms, supra note 30, at 166 (describing some of the investment motivations of vulture funds).
73 See, e.g., Eva Laureano, P.R. en la mesa con los “buitres”, la nueva realidad política, NOTICEL (Jan. 6, 2015), http://www.noticel.com/noticia/170291/p-‐r-‐en-‐la-‐mesa-‐con-‐los-‐buitres-‐la-‐nueva-‐realidad-‐politica.html; Joanisabel González, Inmenso desafío de renegociar, EL NUEVO DÍA (Feb. 3, 2015), http://www.elnuevodia.com/noticias/locales/nota/inmensodesafioderenegociar-‐2003064/; José L. Carmona, Hedge-‐Fund Investors to Square off against Traditional P.R. Bondholders, CARIBBEAN BUS.,
Núm. 4 (2015) COLLECTIVE ACTION CLAUSES FOR PUERTO RICAN BONDS 1209
financial crisis must have played a pivotal role in the broadening of concerns over a sovereign’s creditworthiness. It is probable that it also sparked awareness that remedial government actions, such as austerity measures, bailouts and third-‐party default guarantees, can impose considerable costs on taxpayers. While C.A.C.s are not the ultimate solution to debt crises, they provide debtors and creditors a palliative tool to minimize the transaction costs, reinforce bar-‐gaining rights and make debt workouts more palatable.
Nevertheless, the use of C.A.C.s entails a slew of concerns concerning debt-‐ors that cannot be ignored.74 First, because C.A.C.s reduce ex-‐ante the costs of engaging in debt restructuring, creditors worry that making it easier for sover-‐eigns to obtain debt relief creates a potential moral hazard.75 Governments may invoke C.A.C.s solely to divert capital to politically sweeter alternatives —building a brand new highway instead of honoring an amortization schedule— or they may lose the incentives to control fiscal profligacy and adopt corrective policies. Second, issuers may fall prey to a prisoner’s dilemma tied to a desire to protect their reputation, fearing that a decision to include C.A.C.s may send an unwarranted distress signal to the market.76 Bond terms can also be negatively contrasted to other sovereign issuers on the basis that comparable issuers lack C.A.C.s.77 Intuitively, this preoccupation is entirely rational. The debt securities market is a competitive market; investment decisions respond to particular in-‐terests and no sensible investor would lightly accept a heightened risk of losses without receiving commensurate compensation. However, these signaling con-‐cerns have been assuaged in part by the success of the official sector in promot-‐ing their widespread adoption.78
Third, when sovereigns have multiple debt issuances outstanding, inter-‐creditor equity may be undermined if certain bond series vote in favor of amendments that other series reject. A free-‐rider problem is created when bondholders no longer need to compromise after other bondholders grant relief
Mar. 13, 2014, available at http://www.caribbeanbusinesspr.com/prnt_ed/hedge-‐fund-‐investors-‐to-‐square-‐off-‐against-‐traditional-‐p.r.-‐bondholders-‐9619.html.
74 See Kenneth Kletzer, Resolving Sovereign Debt Crises with Collective Action Clauses, FED. RESERVE BANK OF S.F. (Feb. 20, 2004), http://www.frbsf.org/economic-‐research/publications/ economic-‐letter/2004/february/resolving-‐sovereign-‐debt-‐crises-‐with-‐collective-‐action-‐clauses/.
75 Id.
76 Joy Dey, Collective Action Clauses: Sovereign Bondholders Cornered?, 15 L. & BUS. REV. AMERICAS 485, 513-‐14 (2009). “Introducing such provisions in jurisdictions that have not seen them might send a wrong signal to the market indicating that the issuer already contemplates defaulting on its debt obligations in the future.” Id.
77 Kletzer, supra note 74. “One question . . . is whether countries with predominantly . . . [unani-‐mous action clause] debt will want to issue new bonds with CACs.” Id.
78 As was the case of Mexico in 2003, the European Union in 2012 and the new I.M.F. proposals adopted by Mexico and Kazakhstan which are discussed throughout this article. However, it may be hard for the Federal Government to order the inclusion of C.A.C.s within state and municipal bonds to the extent that doing so tramples over sovereignty and federalism boundaries. Nevertheless, noth-‐ing bars official sector entities such as the United States Treasury, or the Federal Reserve, from adopting an open policy endorsing their adoption within the municipal debt market.
1210 REVISTA JURÍDICA UPR Vol. 84
first; they gain on account of the other bondholder’s loss either because they have stronger leverage over a debtor that has been granted relief by other parties or the relieved debtor has more incentives to satisfy the demands of a discrete minority of holdouts.79 The solution is incorporating into C.A.C.s an aggregation mechanism.80 If different bond issuances were aggregated for the purposes of allowing a restructuring to go forward, holdouts would have to incur greater costs in order to obtain blocking positions within multiple individual issuances.81
It is therefore unsurprising that C.A.C.s have been at the forefront of debates on reforming sovereign debt workouts. They bear upon investor considerations of risk, moral hazard, reputation, equitable treatment and efficiency. Parallel to their growing acceptance, scholars have sought to understand the role C.A.C.s play in either increasing or decreasing an issuer’s cost of borrowing. The under-‐lying assumption these scholars make is that bondholders are risk-‐averse inves-‐tors who require a premium when their return on assets is at risk. In contrast to sovereign bonds, corporate bonds have been a continuous staple of the capital markets and therefore provide a valuable database to study how investors price bond terms. Since Clifford Smith and Jerold Warner published their Costly Con-‐tracting Hypothesis,82 corporate bond covenants have been understood in light of agency costs.83 Restrictive bond covenants improve the cost of capital (i.e., borrowing costs) because investors benefit from an extra layer of protection where there is a need to curb management misbehavior. But this is not always the case. If a debtor’s financial success relies on affording management flexibil-‐ity, restrictive covenants impose agency costs by cabining valuable deci-‐sion-‐making discretion.84 Although a corporate analogy to sovereign debtors is not perfect, the behavioral insights of this hypothesis were the cornerstone that sovereign debt studies built upon: the incentives produced by some contractual terms may influence an issuer’s cost of capital.85
79 See, e.g., Christian Hofmann, Sovereign-‐Debt Restructuring in Europe under the New Model Collective Action Clauses, 49 TEX. INT’L L.J. 385, 397-‐99 (2014); Mark L. J. Wright, Sovereign Debt Restructuring: Problems and Prospects, 2 HARV. BUS. L. REV. 153, 174 (2012); Gelpern & Gulati, supra note 54, at 1639.
80 For example, Greece had thirty-‐six English Law bonds that incorporated C.A.C.s. An inability to aggregate these bonds allowed some bondholders to reject the proposed restructuring and only seventeen were restructured. As a result, Greece had a very small patch of resisting holdouts that eventually received full payment. See Sean Hagan, Debt Restructuring and Economic Recovery, in SOVEREIGN DEBT MANAGEMENT, supra note 35, at 359, 383.
81 To be sure, stronger aggregation mechanisms create their own bundle of moral hazard con-‐cerns.
82 See Clifford W. Smith, Jr. & Jerold B. Warner, An Analysis of Bond Covenants, 7 J. FIN. ECON. 117 (1979), available at http://www.simon.rochester.edu/fac/warner/Jerry%20Papers/JFE-‐June%2079.pdf. See also Choi et al., Pricing Terms, supra note 30, at 168 (discussing the influence of Smith and Warn-‐er’s theory on the study of bond covenants).
83 Choi et al., Pricing Terms, supra note 30, at 168.
84 Id.
85 Id. at 168-‐72 (extrapolating Smith and Warner’s hypothesis and applying it to sovereign debt).
Núm. 4 (2015) COLLECTIVE ACTION CLAUSES FOR PUERTO RICAN BONDS 1211
Notwithstanding this underlying theory, isolating the effects of a single bond term is inherently difficult. Individual provisions are part of a universe of terms that work in tandem with each other. Investors may not even care about bond terms unless they are given a reason to worry that the terms represent a risk to their investments that they’d rather not take. Therefore, potential borrowing costs are important, but do not complete the spectrum of short, medium and long-‐term considerations a policymaker needs to keep in mind to understand the benefits of C.A.C.s. The fact that some of these considerations are specula-‐tive, and dwell on the unknown future, makes the task undoubtedly harder. As a result, focusing solely on potential borrowing costs could do a disservice to those who have a stake in Puerto Rico’s economic health.
ii. The Implications of Sovereign Immunity
Some commentators believe that collective action problems are not as per-‐vasive within the state and municipal debt context as they are with sovereign debt.86 They argue that a state’s ability to invoke sovereign immunity may blunt these problems in various ways. In other words, sovereign immunity trumps any benefit C.A.C.s provide. As a matter of Law, sovereign immunity derives from two sources: the Eleventh Amendment and Common Law.87 While Eleventh Amendment sovereign immunity is a creature of our federalist system of gov-‐ernment,88 Common Law sovereign immunity evolved from a belief that states retained residual sovereign powers when they joined the Union.89 Both doctrines stand for the general proposition that, unless abrogated by contract or Federal Law, a state cannot be sued without its consent. Sovereign immunity’s strongest
86 See, e.g., Skeel, supra note 24, at 683 (“Whatever collective action problems a state faces are quite limited.”); Anna Gelpern, Bankruptcy, Backwards: The Problem of Quasi-‐Sovereign Debt, 121 YALE L.J. 888, 932 (2011-‐2012) (“Quasi-‐sovereigns have advantages over private debtors and some sovereigns: they are substantially sheltered from litigation and attachment and have a powerful counter to creditor collective action problems in the form of sovereign immunity.”). But see Schwarcz, supra note 26, at 327-‐29 (discussing the creditor-‐holdout problems states potentially face).
87 The Eleventh Amendment provides that “[t]he Judicial power of the United States shall not be construed to extend to any suit in law or equity, commenced or prosecuted against one of the United States by Citizens of another State, or by Citizens or Subjects of any Foreign State.” U.S. CONST. amend. XI. Although it can be waived contractually or by federal law, it protects states from being sued by bondholders seeking to recover monetary claims in federal courts. Common Law sovereign immunity emanates from the residual power states inherently retained, which they had as sovereigns before the formation of, or before they joined, the Federal Union. A state’s Common Law sovereign immunity entails that states cannot be sued in their own courts under their own laws, unless they consent via legislation or contract. See Ngiraingas v. Sánchez, 495 U.S. 182, 205 (1990) (Brennan, J., dissenting).
88 See generally JOHN V. ORTH, THE JUDICIAL POWER OF THE UNITED STATES: THE ELEVENTH AMENDMENT IN AMERICAN HISTORY (1987).
89 See supra text accompanying note 87 (discussing Common Law sovereign immunity as being a residue of sovereignty).
1212 REVISTA JURÍDICA UPR Vol. 84
bite comes from its protection against any lawsuit that would ultimately require reaching into state funds for relief.90
Professor Gelpern posits that sovereign immunity may be analogized to Bankruptcy Law’s automatic stay.91 To the extent that a state has not contractual-‐ly waived it, bondholders will find it difficult to turn to the judicial system for remedies. Even with the specter of sovereign immunity removed, bondholders would still likely find limited avenues for recovery from a state facing a dire fiscal crisis. A state, unlike a corporate entity, cannot be liquidated and many of its assets, like roads, schools and hospitals, are exempt from attachment.92 Courts are likely to consider equity and the social, economic and political repercussions of ordering a state to pay while it faces acute distress.93
Puerto Rico, as an unincorporated federal territory, has yet to hear a firm pronouncement from the Federal Supreme Court on the exact contours of its immunity shield. But the First Circuit94 has consistently held that Puerto Rico enjoys Eleventh Amendment sovereign immunity,95 and the Supreme Court has
90 See, e.g., Ford Motor Co. v. Dep’t of Treasury of Ind., 323 U.S. 459 (1945). “And when the action is in essence one for the recovery of money from the state, the state is the real, substantial party in interest and is entitled to invoke its sovereign immunity from suit even though individual officials are nominal defendants.” Id. at 464. See also John J. Gibbons, The Eleventh Amendment and State Sovereign Immunity: A Reinterpretation, 83 COLUM. L. REV. 1889, 2004 (1983) (“Yet under current eleventh amendment doctrine, federal courts would not be able to hear suits by bondholders against the states. No amount of arid theorizing on the nature of state sovereignty or the amenability of enforcements against state treasuries could justify such a catastrophic result.” (footnote omitted)).
91 See Gelpern, supra note 86, at 902-‐03 (“Immunity acts in important respects like an automatic stay on enforcement, a feature of the U.S. and other bankruptcy regimes, which protects the debtor from lawsuits and protects its assets from seizure.”).
92 See Schwarcz, supra note 26, at 335 (“As a political if not constitutional matter, states cannot be liquidated.”).
93 During the 1970’s, the city of New York faced a grave fiscal crisis and was on the brink of de-‐fault. The state enacted emergency legislation providing the city with a debt restructuring mecha-‐nism. The New York Court of Appeals, the state’s highest court, sided with bondholders and declared the state law unconstitutional because it violated the state constitution’s full faith and credit pledge. However, the Court refused to order the city to pay bondholders. In doing so, the Court recognized the implications of ordering a remedy that pushed the city into bankruptcy. See Flushing Nat’l Bank v. Mun. Assistance Corp., 358 N.E.2d 848, 855 (N.Y. 1976) (“It would serve neither plaintiff nor the people of the City of New York precipitately to invoke instant judicial remedies which might give the city no choice except to proceed into bankruptcy.”). For background information on New York’s fiscal crisis see ROGER DUNSTAN, OVERVIEW OF NEW YORK CITY’S FISCAL CRISIS (1995), available at http://www.library.ca.gov/crb/95/notes/V3N1.pdf.
94 The U.S. Court of Appeals for the First Circuit has appellate jurisdiction over the United States District Court for the District of Puerto Rico.
95 See, e.g., Metcalf & Eddy, Inc. v. P.R. Aqueduct & Sewer Auth., 945 F.2d 10, 11 n.1 (1st Cir. 1991) (“It is settled that Puerto Rico is to be treated as a state for Eleventh Amendment purposes.”), rev ‘d on other grounds, 506 U.S. 139 (1993), remanded, Metcalf & Eddy, Inc. v. P.R. Aqueduct & Sewer Auth., 991 F.2d 935, 939 n.3 (1st Cir. 1993) (“We have consistently treated Puerto Rico as if it were a state for Eleventh Amendment purposes.”). But see Adam D. Chandler, Puerto Rico’s Eleventh Amendment Status Anxiety, 120 YALE L.J. 2183 (2011) (arguing that the First Circuit might be mistaken on Puerto Rico’s having Eleventh Amendment immunity).
Núm. 4 (2015) COLLECTIVE ACTION CLAUSES FOR PUERTO RICAN BONDS 1213
also recognized its Common Law sovereign immunity.96 In the same vein, federal courts have recognized that some, but not all, of Puerto Rico’s public entities can invoke its protection.97 Regardless, there is no solace to be found in sovereign immunity. A state can simply refuse to pay, but sovereign immunity cannot be relied upon to securely bind holdouts that stubbornly sit on their claims.98 Dur-‐ing a stalemate, a state would presumably be forced to engage bondholders in a conciliatory manner to minimize exposure to the market’s wrath, punishing them with increased cost of capital, a tarnished reputation and limited market access. Arkansas, the only state to default during the twentieth century, experi-‐enced a similar ordeal and is a case in point.
During the Great Depression, Arkansas’s coffers suffered from revenue shortfalls that made the servicing of highway revenue bonds unsustainable. Faced with a shrinking economy that disabled an ability to raise taxes any fur-‐ther,99 Arkansas considered ways to avoid a default by enacting refunding legisla-‐tion that extended maturities and converted revenue bonds into general obliga-‐tion debt.100 Bondholders succeeded in getting a federal court to enjoin the en-‐forcement of the statute,101 but were ultimately unsuccessful in getting the State to pay up: the State and its treasury enjoyed sovereign immunity.102 The stale-‐
96 People of Porto Rico v. Rosaly y Castillo, 227 U.S. 270, 273 (1913) (“It is not open to controversy that . . . the government which the organic act established in Porto Rico is of such nature as to come within the general rule exempting a government sovereign in its attributes from being sued without its consent.”).
97 For example, federal courts have already determined that the P.R.E.P.A., the P.R.A.S.A. and the P.R. Highway and Transportation Administration do not have sovereign immunity. See Redondo Constr. Corp. v. P.R. Highway & Transp. Auth., 357 F.3d 124, 129 (1st Cir. 2004) (holding that the Puerto Rico Highway and Transportation Authority does not have sovereign immunity); Metcalf & Eddy, Inc. v. P.R. Aqueduct & Sewer Auth., 991 F.2d 935, 939 n.3 (1st Cir. 1993) (holding that P.R.A.S.A. does not have sovereign immunity); Riefkohl v. Alvarado, 749 F. Supp. 374, 376 (D.P.R. 1990) (holding that P.R.E.P.A. does not have sovereign immunity). On the other hand, federal courts have determined that the P.R. Ports Authority does have sovereign immunity. See P.R. Ports Auth. v. Fed. Mar., 531 F.3d 868, 881 (D.C. Cir. 2008) (holding that the P.R. Ports Authority does have sover-‐eign immunity).
98 In her article, Anna Gelpern argues the following:
Although immunity cannot bind holdouts as bankruptcy might, the holdout has few rem-‐edies if the state refuses to pay on the old obligation. However, to the extent a debt re-‐structuring frees up cash flows for payment, some creditors may hold out in hope that a state would choose to repay for reputational or other reasons, even if it cannot be com-‐pelled to do so.
See Gelpern, supra note 86, at 905.
99 See Damon A. Silvers, Obligations without the Power to Fund Them, in WHEN STATES GO BROKE, supra note 4, at 40, 46 (discussing the state’s inability to raise further revenue through taxes).
100 Lee Reaves, Highway Bond Refunding in Arkansas, 2 ARK. HIST. Q. 316, 318 (1943).
101 See Hubbell v. Leonard, 6 F. Supp. 145 (E.D. Ark. 1934) (enjoining an Arkansas state officer from diverting proceeds from taxes that were contractually pledged to pay interest and principal on highway revenue bonds).
102 Id. The refunding law was declared unconstitutional and therefore the state officer could not enforce its provision because unconstitutional legislation takes away the sovereign immunity. But,
1214 REVISTA JURÍDICA UPR Vol. 84
mate was resolved once Arkansas, still crushed by debt, enacted a second re-‐funding law fruit of negotiations with a committee of bondholders. Bondholders “understood that . . . [to keep seeking judicial remedies] would be futile in a con-‐text in which Arkansas’s economy had essentially collapsed.”103
Some could argue that this episode illustrates how sovereign immunity inev-‐itably leads to the same results C.A.C.s attempt to provide: a negotiated solution. However, this idea proves illusory after considering the reputational and transac-‐tion costs incurred by defaulting and prolonging a conflict with bondholders. If Arkansas wanted to tap the debt markets, it would have to account to potential investors for its refusal to pay, thereby undermining any bargaining power it might otherwise have to negotiate. C.A.C.s, unlike a default with recourse to immunity, avoid such an awkward situation. Avoiding a default allows an issuer to save face, preserve the market’s good will and dissuades bondholders from seeking judicial remedies. Thus, “CACs . . . [can still] be seen as a potential solu-‐tion to the creditor-‐holdout problem of states” notwithstanding their ability to invoke sovereign immunity.104 The palpable benefit that C.A.C.s provide, that sovereign immunity does not, lies in providing for a more cost-‐efficient, sleeker procedure. Arguments in favor of relying on sovereign immunity, instead of C.A.C.s, to bind dissenting bondholders underestimate the benefit of a higher likelihood of averting a default and inappropriately simplify the inherent costs of any debt workout.
iii. Federal and State Law Considerations
There is no federal legal impediment for Puerto Rico and other U.S. states to include C.A.C.s in the terms of their debt issuances. Experts on the matter have testified before the Securities and Exchange Commission that C.A.C.s may solve a state and municipal issuer’s new economic reality to the extent that they do not foreclose market access.105 Experts also invite a careful balancing of the pros and cons of C.A.C.s prior to a decision to include them. For instance, the utility of C.A.C.s depends on using them constructively, rather than using them to avoid “financial responsibility or denying the benefit of the bargain.”106 This caveat voices the same concerns of moral hazard and market acceptance emerging prior to Mexico’s move in 2003. That they have not become a staple clause in state and municipal bonds may therefore reflect a prisoner’s dilemma instead of a down-‐
the remedy sought by bondholders ultimately hinged on repayment from state funds, which still enjoyed sovereign immunity protection.
103 Silvers, supra note 99, at 47.
104 Schwarcz, supra note 26, at 329 (discussing C.A.C.s as a way to tackle coordination problems states could face).
105 JAMES E. SPIOTTO, FIELD HEARING ON THE STATE OF THE MUNICIPAL SECURITIES MARKET: DISTRESSED COMMUNITIES 5-‐6 (2011), available at http://www.sec.gov/spotlight/municipalsecurities/statements072911/spiotto.pdf.
106 Id. at 6.
Núm. 4 (2015) COLLECTIVE ACTION CLAUSES FOR PUERTO RICAN BONDS 1215
right market rejection, with no single state wanting to take the initial step into the uncertainty of contractual innovation.
And yet, even though C.A.C.s are legal under federal laws, they still need to fit within the existing parameters of state laws. Some states have granted certain bondholders additional protections in the form of constitutional guarantees that undermine the utility of C.A.C.s.107 Puerto Rico is no exception. The Common-‐wealth’s constitutional guarantees relating to its general obligation debt have traditionally been perceived to place strong protective barriers around bond-‐holders.108 Section 8 of article VI of the Commonwealth’s constitution provides that if available revenues are insufficient to meet appropriations for that year, “interest on the public debt and amortization thereof shall” receive first priority in an order of disbursements established by law.109 In the event that section 8 of article VI is triggered, section 2 of article VI stipulates that “[t]he Secretary of the Treasury may be required” by a court acting at the behest of a bondholder to ap-‐ply any available revenues towards the payment of interest and amortization of the public debt.110
Some commentators argue that these types of constitutional guarantees may simply require that payments “go first to bonds, not that the principal amount of the bonds be untouched.”111 Because C.A.C.s are a contractual term subject to Contract Law, courts can find a hook to enforce them as a matter of upholding a bargain. But the constitutional priority guarantees may nevertheless hinder a C.A.C.’s ability to induce a compromise between general obligation bondholders and the Commonwealth by exacerbating collective action problems. Bondhold-‐ers may feel no impetus whatsoever to grant relief with the knowledge that they were promised a payment in accordance with a priority enshrined in a constitu-‐tion. If one were to analogize the constitutional debt priority provisions under a C.A.C.-‐regime as an allocation of an entitlement in favor of bondholders,112 the
107 See Skeel, supra note 24, at 694-‐97 (discussing how some state constitutions allocate priorities over revenue streams).
108 See, e.g., Mary Williams Walsh & Michael Corkery, Puerto Rico Wants to Incur More Debt to Regain Financial Footing, N.Y. TIMES (Feb. 18, 2014, 9:11 PM), http://dealbook.nytimes.com/2014/02/18/puerto-‐rico-‐wants-‐to-‐incur-‐more-‐debt-‐to-‐regain-‐financial-‐footing/?_r=0 (“Even though the constitution explicitly gives general-‐obligation bonds priority over all other expenditures, that provi-‐sion has not been tested in court.”); John Marino et al., Prepa is the Problem, CARIBBEAN BUS., July 10, 2014, available at http://caribbeanbusinesspr.com/prnt_ed/prepa-‐is-‐the-‐problem-‐10138.html.
109 CONST. PR art. VI, § 8.
110 Id. § 2 (emphasis added). This provision may also be construed as a waiver of immunity. See supra Part I.C.ii.
111 See Skeel, supra note 24, at 695 n.81 (discussing an analogous provision in the California Con-‐stitution).
112 The New York Court of Appeals’ decision in Flushing Nat’l Bank v. Mun. Assistance Corp. illus-‐trates how a court could interpret the Puerto Rican constitution’s debt priority provisions. Flushing Nat’l Bank v. Mun. Assistance Corp., 358 N.E.2d 848 (N.Y. 1976). See Clayton P. Gillette, Bondholders and Financially Stressed Municipalities, 39 FORDHAM URB. L.J. 639, 648 (2011-‐2012) (“What the . . . [court] gave with one hand, however, it withdrew with the other. The court ostensibly did not want
1216 REVISTA JURÍDICA UPR Vol. 84
likely end result might mean that in any negotiations with general obligation bondholders, the deck is stacked against the Commonwealth. Perhaps this ne-‐cessitates C.A.C.s that are specifically tailored for these bonds, which would maximize bondholder leverage. This could be achieved by foregoing aggregation mechanisms and requiring a particularly higher-‐than-‐normal consent threshold in order to preserve the constitutional guarantees within the architecture of the C.A.C. mechanism itself. This could also entail that the Commonwealth would need to approach a debt workout with a credibility-‐enhancing strategy. For ex-‐ample, the onus of transparency may well be on the Commonwealth to prove that it is making a request for relief after exhausting alternative fiscal remedies that impose minimal losses on bondholders.
The following section will survey seven empirical studies on the link be-‐tween C.A.C.s and borrowing costs for sovereign debtors. Recent events should bear upon how we understand them in the Puerto Rican context, particularly the Greek debt crisis and the ongoing theatrics surrounding the Argentine bond-‐holder litigation. The financial media has become well acquainted with sover-‐eign debt crises where debtors and bondholders have endured the detrimental effects of a messy default. The media has also continuously covered state fiscal crises and particularly emphasized the prospect of a Puerto Rican default akin to Argentina or Greece.113 Since modern capital markets have cut their teeth when it comes to sovereign debt restructurings and messy defaults, modern studies are likely to provide a better picture of how the inclusion of C.A.C.s is perceived by market participants, and may influence the Commonwealth’s borrowing costs. Nevertheless, it is important to note that some authors of these studies caution against interpreting their results as conclusive indicators of what to expect.
I I . SURVEY OF STUDIES ON BORROWING COSTS FROM USING C.A.C.S
A. Early Studies: Eichengreen & Mody (2000 and 2004), Gugiatti & Richards (2003), Becker, Richards & Thaicharoen (2003)
In 2000, Barry Eichengreen and Ashoka Mody were among the first to study the link between C.A.C.s and borrowing costs.114 At the time, only English Law bonds systematically included C.A.C.s; the Argentine messy default had yet to occur and Mexico’s inclusion of C.A.C.s in New York Law bonds was still on the horizon. Eichengreen and Mody used primary-‐market data of emerging market bonds (i.e., yield spreads at their issuance date) issued from 1991 to 1998, segre-‐gated based on their credit rating. To isolate the effects of C.A.C.s, they attempt-‐
to authorize creditors to bankrupt the city in order to obtain payment. Once it articulated a clear entitlement in favor of creditors, the court expressed an unwillingness to enforce it.”).
113 See articles cited in supra note 14 (discussing sovereign debt analogies to Puerto Rico).
114 See Barry Eichengreen & Ashoka Mody, Would Collective Action Clauses Raise Borrowing Costs? (Nat’l Bureau of Econ. Research, Working Paper No. 7458, 2000) [hereinafter Eichengreen & Mody, Would Collective Action Clauses], available at http://www.nber.org/papers/w7458.pdf.
Núm. 4 (2015) COLLECTIVE ACTION CLAUSES FOR PUERTO RICAN BONDS 1217
ed to limit data distortions from choice of law provisions and changes in the pool of borrowers by accounting for endogeneity (when a variable is correlated with the results). They concluded that C.A.C.s reduced the borrowing costs for credit-‐worthy issuers while, if anything, minimally increasing the costs for less credit-‐worthy issuers.115 The spread differential was due to particular behavioral incen-‐tives tied to credit rating of an issuer’s sovereign state: issuers with strong credit value their ability to tap the debt markets, minimizing a potential source of mor-‐al hazard, and therefore benefit from being able to “avail themselves of [an] or-‐derly restructuring.”116 According to them, the market believes that the weaker the credit rating the stronger a moral hazard that increases borrowing costs. But the fact that C.A.C.s allow for orderly restructurings counter-‐balances this effect. An ability to negotiate debt relief in an orderly manner is valued by creditors, and renders the cost of capital impact of moral hazard insignificant.117
A subsequent 2004 update from the same authors obtained similar results, but observed stronger borrowing cost differentials tied to credit rating.118 This update, like the original study before it, concluded that C.A.C.s reduce the bor-‐rowing costs for credit-‐worthy issuers, but instead found a sizable effect for issu-‐ers from weak credit countries: “[The point] estimates suggest that issuers from countries . . . [with low credit ratings] pay about a 32% premium for the choice of UK law [with imbedded C.A.C.s].”119 Like their prior study, they argued that cre-‐ditworthy issuers enjoyed lower spreads because they had the option of an or-‐derly restructuring. Yet the advantage of an orderly restructuring for weaker credit issuers is “offset by the moral hazard and additional default risk associated with the presence of renegotiation-‐friendly loan provisions.”120 However, the authors make a poignant observation. A form of reverse-‐moral hazard exists when an official sector entity influences an issuer’s debt financing. A lack of C.A.C.s “encourages reckless lending and reckless borrowing” because the official sector prefers to provide liquidity injections to avoid the costs of a default.121 An endless death spiral ensues when debt incurrence prolongs the overhang prob-‐lem without enabling a final solution.
This final observation plausibly rings true in the municipal debt context, even if it’s somewhat farfetched. Perhaps to protect the integrity of the munici-‐pal debt markets, or to control and confine the contagion effect from spilling over to other states’ debt, the Federal Government would feel compelled to pro-‐vide liquidity injections to a state when it’s looking down the precipice. C.A.C.s
115 Id. at 4-‐5.
116 Id. at 4.
117 Id. at 19.
118 See Barry Eichengreen & Ashoka Mody, Do Collective Action Clauses Raise Borrowing Costs?, 114 ECON. J. 247, 262 (2004) [hereinafter Eichengreen & Mody, Do Collective Actions].
119 Id. at 258.
120 Id. at 262 (emphasis added).
121 Id. at 263.
1218 REVISTA JURÍDICA UPR Vol. 84
might minimize a moral hazard derived from an expectation to be rescued by a third-‐party, to the extent that they avoid outside intervention and the solution to state debt problem is left to states themselves.
In 2003, two follow-‐up studies were conducted in the wake of debates on re-‐forming sovereign bond contracts. Becker, Richards and Thaicharoen were the first to incorporate secondary market data (bond spreads after the issuance and when traded freely amongst investors).122 They argued that secondary market data corrects endogeneity distortions and allows for event-‐based measurements. The effect of C.A.C.s can therefore be measured with greater accuracy in light of the market’s perception at specific point in time.123 They also limited their sam-‐ple base to the period between 1998 and 2001. The choice of time period is im-‐portant because it excludes the period where a debate over the merits of the inclusion of C.A.C.s occupied a prominent position. There was “no evidence to support the proposition that the use of CACs has increased borrowing costs for lower-‐rated issuers . . . .”124 —supporting the market’s “conventional wisdom . . . [that] most market participants . . . [are] unaware of this aspect of the legal doc-‐umentation . . . . [This suggests that C.A.C.s are] not associated . . . with . . . mor-‐al hazard”.125 However, they caution against using their results as a definitive guide to the future,126 and emphasize that they only reflect the impact of C.A.C.s as of a specific moment in time. In describing the market’s conventional wisdom, they make implicit observations on how risk perceptions are malleable:
The sell-‐side research of investment banks never refers to CACs as a factor in explaining the pricing of individual bonds. The several data and news services that report in detail on each new issue never explain the pricing of a new bond in terms of the presence or absence of CACs. Further, many borrowers switch frequently between governing laws—apparently sometimes without being aware of it—which seems inconsistent with a careful consideration of the benefits of a lower yield versus the ability to easily restructure when it suited them. Finally, bond ratings from ratings agencies do not differ based on governing law—agencies have not considered governing law as a risk factor.127
Many prospective bondholders rely on market analysts to digest bond terms and elucidate their implications. Conventional market wisdom on moral hazard
122 See Torbjörn Becker et al., Bond Restructuring and Moral Hazard: Are Collective Action Clauses Costly?, 61 J. INT’L ECON. 127, 129 (2003) (“The main innovation of this paper is that it is the first to do a systematic study of the secondary market yields of a large sample of bonds.”).
123 See id. (“Secondary market data allow the researcher to analyze the pricing of a large number of existing bonds at particular points in time, including before and after different events that may have resulted in changes in the relative values that investors place on bonds with and without CACs.”).
124 Id.
125 Id. at 130.
126 Id. at 159.
127 Id. at 158 (footnote omitted).
Núm. 4 (2015) COLLECTIVE ACTION CLAUSES FOR PUERTO RICAN BONDS 1219
vis-‐à-‐vis C.A.C.s is therefore subject to the changing risk perceptions of certain market actors, and having witnessed or experienced the effects of a debt crisis or default may influence how these perceptions evolve. Becker, Richards and Thaicharoen recognize the influence of the then ongoing reform debates and Argentina’s 2002 default as possibly transforming bondholder expectations.128 By way of example, a sovereign bond’s governing law has become an important el-‐ement in dictating restructuring outcomes. The fact that Argentine bonds were subject to New York Law has been the reason why United States courts continue to provide a forum to entertain holdout claims. Similarly, Greece was able to restructure a substantial portion of its outstanding debt in part because Greek Law governed vast swaths of Greek debt. This permitted Greece to use governing law to its advantage to legislate a retroactive C.A.C. into outstanding bond terms and facilitate a subsequent exchange offer.129 To be sure, the importance of gov-‐erning law has recently resurfaced in the Puerto Rican context. In March of 2014, one of the principal issues hovering over a Commonwealth general obligation bond issuance was whether New York or Puerto Rico Law would govern —New York prevailed, the bonds were issued under New York Law.130
Mark Gugiatti and Anthony Richards also conducted a study, in which they concluded that C.A.C.s have negligible impact on bond prices.131 Their study used data from January 31, 2003, when the contract reform debates would have pre-‐sumably influenced market perceptions. They argue that their results indicate that the market had “not focused on which bonds had CACs.”132 Nor did they appear relevant to pricing bonds notwithstanding a public debate on including C.A.C.s as a specific remedy to creditor coordination problems.133 They add a caveat, however, by noting that Mexico successfully incorporated C.A.C.s in a bond issuance on February 26, 2003, followed by other noteworthy sovereign issuers. The primary yield of these bonds was similar to prior issuances lacking C.A.C.s, which suggested to them that U.S. investors perceived these clauses as
128 See id. at 159 (“[T]he default of Argentina and the debate in early 2002 over improving the mechanisms for dealing with sovereign debt problems might imply a regime change that makes the earlier results less relevant.”).
129 Lee C. Buchheit & G. Mitu Gulati, How to Restructure Greek Debt 10-‐13 (Duke Law Scholarship Repository, Working Paper No. 47, 2010) [hereinafter Buchheit & Gulati, How to Restructure], availa-‐ble at http://scholarship.law.duke.edu/cgi/viewcontent.cgi?article=2959&context=faculty_scholarship.
130 Felix Salmon, Why Puerto Rico’s Bonds are Moving to New York, REUTERS (Mar. 3, 2014), http://blogs.reuters.com/felix-‐salmon/2014/03/03/why-‐puerto-‐ricos-‐bonds-‐are-‐moving-‐to-‐new-‐york/.
131 Anthony Richards & Mark Gugiatti, Do Collective Action Clauses Influence Bond Yields? New Evidence from Emerging Markets, 6 INT’L FIN. 415, 418 (2003) (“The results for both dates suggest that there is still little convincing evidence that CACs have had an economically or statistically significant impact upon bond pricing in the secondary market.”).
132 Id. at 443.
133 Id. at 417-‐18.
1220 REVISTA JURÍDICA UPR Vol. 84
congruent with their bundle of creditor’s rights and had no impact on borrowing costs.134
B. Recent Studies: Choi, Gulati & Posner (2011), Bardozzetti & Dottori (2013), Bradley & Gulati (2013)
Recent studies have benefitted from a radically altered landscape. Mexico’s adoption of C.A.C.s proved to be a positive signaling event that eliminated the prisoner’s dilemma among sovereign issuers, and opened the floodgates for wid-‐er C.A.C. acceptance in the sovereign debt markets.135 “[I]n the space of just 19 months, the inclusion of CACs in sovereign bonds issued under New York law . . . switched from being the exception to becoming the market standard,”136 anointed with a certain gravitas. Moreover, and as mentioned before, the Argen-‐tine holdout saga and the Greek debt crisis brought back to the limelight the important role C.A.C.s have in facilitating debt restructurings.
In 2011, in the wake of the Eurozone debt crisis, Stephen Choi, Mitu Gulati and Eric Posner set out to test a hypothesis that boilerplate bond terms are ig-‐nored by investors and have no influence on how bonds are priced. They weaved into this debate the fact that Germany was leading the calls for an orderly mech-‐anism that minimized the need for bailouts with taxpayer money.137 In other words, the European official sector was beginning to openly endorse a notion that private creditors had to accept a share of the burden in helping countries regain financial footing. This was significant to the extent that bondholder losses were no longer a matter of if, but a question of when. As one would expect, the Greek crisis was fertile ground to measure how bondholders responded to this shift in expectations.
Choi, Gulati and Posner used governing law, not C.A.C.s, as an indicator of how bondholders priced contractual protections. Specifically, they wanted to measure whether or not the market was pricing the holdout premium prior to realizing Greece’s distress and identifying if and when the market recognized the implications of having Greek bonds governed by a particular legal system, as-‐suming that no price differences existed prior to this date.138 Greek bonds were traditionally issued under either Greek or English Law. Greek Law bonds were assumed to be riskier because they could be unilaterally restructured through
134 Id. at 444 (“By almost all accounts, the inclusion of CACs has had no impact on the yields paid by these borrowers, with many market participants indicating that the inclusion of CACs has been no more than a footnote to the issues.”).
135 See JOHN DRAGE & CATHERINE HOVAGUIMIAN, COLLECTIVE ACTION CLAUSES (CACS): AN ANALYSIS OF PROVISIONS INCLUDED IN RECENT SOVEREIGN BOND ISSUES 2-‐3 (2004) (discussing how many sover-‐eign issuers, particularly emerging market countries, expeditiously included C.A.C.s in their issuanc-‐es following Mexico decision to do the same).
136 Id. at 3 (emphasis omitted).
137 Choi et al., Pricing Terms, supra note 30, at 164.
138 Id. at 167.
Núm. 4 (2015) COLLECTIVE ACTION CLAUSES FOR PUERTO RICAN BONDS 1221
legislative action by the Greek government. In contrast, bonds issued under Eng-‐lish Law incorporated C.A.C.s whereby Greece had to negotiate with creditors for relief because they were subject to a framework over which Greece had little to no influence.139 Greece’s English Law bonds thus represented an opportunity to gauge the existence of a greater holdout premium.
Governing law represents a different, albeit presumably stronger, set of bondholder rights than a C.A.C.140 Using governing laws as a proxy for holdout premium may not be all that different from an analogous focus on C.A.C.s. In lay terms, a holdout premium is the amount an investor is willing to forgo in higher returns because they contract for stronger rights ex-‐ante to a default.141 As ana-‐lyzed by Choi, Gulati and Posner, governing law and C.A.C.s elicit the same un-‐derlying worry from bondholder creditors: the ease and ability with which a sov-‐ereign issuer can alter the payment terms of an outstanding debt instrument. Using data on bond yields for outstanding English Law Greek bonds, they found evidence that a small holdout premium existed prior to November 2009. Not only did this evidence imply that the market may have recognized holdout rights ex-‐ante, but it departed from earlier studies “which . . . focus[ed] . . . on the ex post costs of delays to restructurings as a result of holdout litigation.”142
The holdout premium increased once the market acquainted itself with the financial distress of the sovereign debtor —in other words, the yield on Greek Law bonds spreads tended to increase as the market digested knowledge of Greece’s ailing health, while the English Law bond yields remained roughly the same. Recognizing the difficulty of making accurate deductions from the evi-‐dence before them, the authors opted for a conservative explanation: the hold-‐out premium appears to be valuable to creditors planning to resist restructuring attempts.143 This reflects the possibility that English Law Greek bond investors made a conscious trade-‐off where they would accept a lower yield in return for
139 Id. at 165 (“In contrast, with the small fraction of English-‐law bonds, Greece had relatively little that it could do other than hope to enter into a restructuring agreement with its creditors . . . .”).
140 In their work, Lee C. Buchheit and G. Mitu Gulati posit that:
International investors are often leery of buying debt securities of emerging market sover-‐eign issuers that are governed by the law of the issuing state. Why? Because investors fear that the sovereign might someday be tempted to change its own law in a way that would impair the value or the enforceability of those securities. Such changes in local law would normally be respected by American and English courts . . . .
See Buchheit & Gulati, How to Restructure, supra note 129, at 10.
141 See generally Choi et al., Pricing Terms, supra note 30, at 166-‐68 (discussing how investors are willing to forgo a higher yield when they contract for bond terms that make the amendment of pay-‐ment terms presumably harder).
142 Id. at 167.
143 See id. at 186 (“Why should bonds governed by different laws be priced differently if they will be treated the same in a restructuring? The answer is that the contract protections are valuable to . . . bondholders who plan to hold out . . . .”).
1222 REVISTA JURÍDICA UPR Vol. 84
greater protections against a default (or unilateral restructuring).144 Nevertheless, it is uncertain how these results can reveal anything applicable to the Common-‐wealth’s context. They arguably suggest that C.A.C.s might lower an issuer’s cost of capital, but the holdout premium depended on investors having been served two options on the table: exposure to a unilateral restructuring145 or the ability to influence the outcome of a restructuring.146 This menu is not available to Com-‐monwealth bondholders since none of the Commonwealth issuances appear to have a collective modification provision akin to a C.A.C. For our purposes, these results might be better understood as suggesting that the ebb and flow of cur-‐rent events has some effect on the risk premium bondholders are willing to pay.
The Greek debt crisis culminated in the largest sovereign debt restructuring to date.147 The European Union subsequently adopted a rule requiring the inclu-‐sion of C.A.C.s in all Euro area sovereign bonds.148 In light of this, Alfredo Bar-‐dozzetti and Davide Dottori, of Italy’s central bank, conducted a study to deter-‐mine the potential effects of this policy.149 They focused on the secondary market for the period between March of 2007 and April of 2011, employed a panel model
144 See id. at 164 (“To the extent English law provides greater protection against a debt restructur-‐ing than Greek law (after all, Greece cannot change English law to suit its purposes), investors pur-‐chasing English-‐law governed bonds made a trade-‐off—accepting lower yields in return for greater protections against default.”). Regardless of whether this is true, during the Greek restructuring, some bondholders successfully rejected proposed amendments. See Hagan, supra note 80.
145 A unilateral restructuring by the Commonwealth is likely impossible. For example, federal and state constitutional provisions cabin a state’s ability to modify the terms of an outstanding contract unless the circumstances make such changes reasonable and necessary. See U.S. Trust Co. v. New Jersey, 431 U.S. 1, 25 (1977) (holding that the Contracts Clause prohibits a state from impairing its obligations unless reasonable and necessary to fulfill an important public purpose that does not merely reflect the preferences of the State to divert contracted funds towards preferable alternatives); CONST. PR art. II, § 7. “No laws impairing the obligation of contracts shall be enacted.” Id. See also Trinidad Hernández v. ELA where the Supreme Court of Puerto Rico summarized their Contracts Clause analysis:
To analyze the constitutional validity of a statute under the impairment of contractual ob-‐ligations clause, the test of reasonableness applies. Thus, in assessing the State’s interfer-‐ence with private contracting, you must first determine if there is a contractual relation-‐ship and if the modification is a substantial or severe impairment. If there is a substantial or severe impairment, the governmental interference is evaluated to see if it responds to a legitimate interest and if it is rationally related to the achievement of the objective.
Trinidad Hernández v. ELA, 188 DPR 828, 834-‐35 (2013) (per curiam) (citations omitted) (translated by author).
146 See Choi et al., Pricing Terms, supra note 30, at 185 (“[E]ven in the English-‐law bonds, there exists a mechanism to quash the holdout.”).
147 See Jeromin Zettelmeyer et al., The Greek Debt Restructuring: An Autopsy 2 (Peterson Inst. for Int’l Econ., Working Paper No. 13-‐8, 2013), available at http://www.iie.com/publications/wp/wp13-‐8.pdf.
148 See generally Hofmann, supra note 79 (discussing the EU’s decision to require members to include C.A.C.s).
149 Alfredo Bardozzetti & Davide Dottori, Collective Action Clauses: How Do They Affect Sovereign Bond Yields?, 92 J. INT’L ECON. 286 (2014).
Núm. 4 (2015) COLLECTIVE ACTION CLAUSES FOR PUERTO RICAN BONDS 1223
approach (measuring yield spreads over a period of time, instead of a specific moment in time), and relied on software indicating the presence or absence of C.A.C.s, and not governing law, as a “proxy to gauge . . . [their] adoption.”150 Re-‐sults suggested that C.A.C.s influenced borrowing costs in a “non-‐linear . . . [U-‐shape] according to the issuer’s rating,” which decreased for issuers hovering in the middle of the credit rating scale while highly and poorly rated issuers saw no effect on their cost of capital.151 Placing the ex-‐post benefits of an orderly restruc-‐turing and moral hazard on opposite sides of a balance, they concluded that mid-‐rated issuers benefit from C.A.C.s. While the risk of default remains, moral hazard does not offset the ex-‐ante benefits of an orderly restructuring. No effect was detected when the issuer had a strong credit rating, but borrowing costs increased for weakly rated issuers because their risk of default is higher and are therefore more prone to deploy C.A.C.s in the near future.152
The latest of these studies, by Michael Bradley and G. Mitu Gulati in 2013, recognized from the onset that the body of empirical research on C.A.C.s and borrowing costs remains ambiguous, in part due to limitations in available da-‐ta.153 An important innovation of this study is a focus on different consent thresholds as a possible cost factor tied to C.A.C.s. They rejected governing law as a proxy, arguing that English and New York Law bonds differ between and within themselves in more substantial ways than just an amendment mecha-‐nism. Their data sample spans the period between January of 1990 and Decem-‐ber of 2010, specifically excluding data from 2011 and 2012 because market senti-‐ment stemming from the Eurozone crisis could have contaminated data by un-‐derpricing emerging market debt and overpricing developed country debt.154 They did not detect evidence that C.A.C.s increase borrowing costs: “Contrary to much of the existing literature, we find that the presence of a CAC in a sovereign bond issue is associated with a lower, not higher, cost of capital, especially for financially weak issuers.”155 They did, however, note a “positive relation between spreads and the number of votes required to change the payment terms.”156 In other words, higher consent thresholds could potentially increase borrowing costs. These impediments, in addition to others such as acceleration clauses, made it difficult to have cost-‐efficient restructurings.157
150 Id. at 287.
151 Id. at 299.
152 Id.
153 Bradley & Gulati, supra note 41, at 2.
154 Id. at 11.
155 Id. at 53.
156 Id.
157 A high threshold could allow a holdout to amass a controlling position to block a restructur-‐ing. See id. (“We find a significant positive relation between spreads and the number of votes re-‐quired to change the payment terms.”).
1224 REVISTA JURÍDICA UPR Vol. 84
Bradley and Gulati recognize that their results present “somewhat of a co-‐nundrum.”158 It was not readily apparent why the European Union needed to take proactive measures compelling members to include C.A.C.s in contrast to their widespread and voluntary use by New York Law issuers.159 After all, effi-‐ciency-‐enhanced C.A.C.s (with lower consent thresholds) appeared to lower the cost of capital.160 The authors hypothesized that the official sector was predicting future debt renegotiations by Eurozone members.161 They also posit that this reflects the reality of issuing debt while being part of a monetary union. Individ-‐ual debtors prefer to be bailed out by other Eurozone members instead of incur-‐ring the reputational costs of restructuring the claims of private creditors. Be-‐cause New York Law issuers do not have similar incentives, and the Federal Gov-‐ernment is not likely going to risk taxpayer money on their behalf, C.A.C.s are therefore the primary tool they have to reduce restructuring costs ex-‐ante.162
IV . THE C .A.C. IN PRACTICE AND RECENT DEVELOPMENTS
A. The Belizean Experience
Prior to 2013, C.A.C.s had been reportedly used in thirty five percent of sov-‐ereign bond exchanges.163 As recently as 2012, eighty three percent of outstanding New York Law sovereign bonds included C.A.C.s, a percentage that is expected to grow as the stock of old debt is retired and new, C.A.C.-‐flavored bonds are introduced.164 Notwithstanding this shift, Belize has been the only country to actually deploy modern C.A.C.s previously included within outstanding New York Law bonds.165 The Belizean experience thus provides useful insight into how
158 Id. (“At first blush, our general results present somewhat of a conundrum.”).
159 Cf. RODRIGO OLIVARES-‐CAMINAL ET AL., DEBT RESTRUCTURING 438-‐39 (2011) (discussing how some commentators believed that countries like Chile included C.A.C.s after being pressured by American officials).
160 Bradley & Gulati, supra note 41, at 53-‐54 (discussing the implications of the results of their research).
161 Id. at 54 (“Our conjecture is that CACs are being mandated by the Official Sector because the officials anticipate (expect) that the bonds of certain nations at least will likely have to be renegotiat-‐ed prior to their maturity . . . .”).
162 Id.
163 See Elena Duggar, The Role of Holdout Creditors and CACs in Sovereign Debt Restructurings, in SOVEREIGN DEFAULTS SERIES 25, 36 (2013), available at https://www.moodys.com/researchdocumentcontentpage.aspx?docid=PBC_150162 (discussing the number of sovereign debt restructurings involv-‐ing the use or activation of C.A.C.s).
164 See Daniel Bases, The Government’s Man When Creditors Bay, REUTERS (May 23, 2012), http://www.reuters.com/article/2012/05/23/us-‐sovereign-‐buchheit-‐idUSBRE84M07N20120523 (“By 2019 just $29 billion in sovereign debt won’t be covered by a collective action clause.”).
165 See INT’L MONETARY FUND, STRENGTHENING THE CONTRACTUAL FRAMEWORK TO ADDRESS COLLECTIVE ACTION PROBLEMS IN SOVEREIGN DEBT RESTRUCTURING 30 (2014) [hereinafter INT’L MONETARY FUND, STRENGTHENING THE CONTRACTUAL FRAMEWORK], available at http://www.imf.org/
Núm. 4 (2015) COLLECTIVE ACTION CLAUSES FOR PUERTO RICAN BONDS 1225
C.A.C.s operate in a market relatively proximate to the municipal debt market. Moreover, it sheds light on how they work in practice and refine our understand-‐ing of how the possibility of increased borrowing costs should be balanced against the proven effect of these clauses in streamlining the restructuring pro-‐cess and averting a default.
Twice in less than ten years, Belize lost the ability to honor its outstanding debt obligations and was forced to restructure its external debt.166 The 2006-‐2007 and 2012-‐2013 restructurings employed the traditional vehicle sovereigns seeking debt relief opt for: an exchange offer. Exchange offers are structured similar to corporate tender offers where the sovereign offers to existing bondholders a new debt instrument in exchange for the instrument they currently hold. Successful exchange offers can provide debt relief by extending bond maturities, reducing par value and lowering interest rates in addition to amending non-‐financial bond terms such as governing law.167 The key, however, is to make an offer sufficiently attractive to induce bondholders to tender the old instrument in exchange for a new instrument that incorporates the debt relief a sovereign needs (the haircut).
For debt exchange offers to be successful, a minimum threshold of tendering bondholders is required.168 In fact, one of the chief objectives behind the design of an exchange offer is to achieve the highest participation possible.169 Neverthe-‐less, every exchange can encounter its fair share of problems and costs. Some bondholders can refuse to accept the offer while they wait to see if the sovereign improves the terms of the offer or agrees to pay them back in full.170 In order to incentivize bondholder participation, an issuer needs to offer both carrots and sticks.171 Carrots might include better legal covenants in the new bonds, a menu of new instruments to choose from and cash payments, all of which incentivize bondholders to tender. In contrast, sticks diminish the value of the old bonds and try to coerce bondholders to tender.172 One stick used in exchange offers is exit consents, another tool in the restructuring arsenal discussed in the next
external/np/pp/eng/2014/090214.pdf (“Since 2003, there have been only two cases in which a CAC was used to restructure a New York law-‐governed sovereign bond (Belize in 2007 and again in 2013), neither of which resulted in a legal challenge.”).
166 See generally Tamon Asonuma et al., Sovereign Debt Restructurings in Belize: Achievements and Challenges Ahead (I.M.F., Working Paper No. 14/132, 2014), available at https://www.imf.org/external/pubs/ft/wp/2014/wp14132.pdf.
167 See OLIVARES-‐CAMINAL ET AL., supra note 159, at 382 (discussing the different ways sovereign debt restructurings facilitate debt relief).
168 Udaibir S. Das et al., Sovereign Debt Restructurings 1950-‐2010: Literature Survey, Data, and Stylized Facts 13 (I.M.F., Working Paper No. 12/203, 2012), available at https://www.imf.org/external/pubs/ft/wp/2012/wp12203.pdf.
169 Id. at 22.
170 See Lee C. Buchheit & G. Mitu Gulati, Exit Consents in Sovereign Bond Exchanges, 48 UCLA L. REV. 59, 64-‐65 (2000-‐2001) [hereinafter Buchheit & Gulati, Exit Consents] (discussing potential bond-‐holder reactions to an issuer’s exchange offer).
171 Das et al., supra note 168, at 21-‐23.
172 Id.
1226 REVISTA JURÍDICA UPR Vol. 84
paragraph.173 Exit consents, however, do not harbor an issuer from holdouts that resist the terms of an exchange offer and threaten to unwind the whole restruc-‐turing. C.A.C.s can be used to compensate for these limitations, neutralizing the free-‐rider effect but not necessarily at the expense of a bondholder’s bargaining leverage. Belize is a case in point.
Belize recognized from the onset that their treatment of bondholders throughout any negotiations would have a direct impact on its market reputa-‐tion, and it opted to engage bondholders with transparency, ultimately achieving an impressive ninety-‐eight percent participation rate.174 This restructuring also resulted in the issuance of a single superbond in exchange for most of the out-‐standing bonds. The high participation rate was facilitated, in part, by using C.A.C.s in conjunction with exit consents.175 Exit consents are particularly useful when restructuring bonds lacking C.A.C.s and allow debtors to circumvent una-‐nimity requirements (i.e., to modify payment terms) by requiring tendering bondholders to vote in favor of amending bond provisions that do not require unanimous approval (i.e., governing law, waiver of immunity, acceleration, etc.). “The tactical objective . . . is to amend the instruments that . . . [would] be left in the hands of holdout creditors after the restructuring . . . in a manner that . . . [makes them] less valuable or more difficult to enforce.”176
For Belize’s first restructuring, the use of exit consents alone resulted in a 96.8% participation rate. However, Belize had previously followed Mexico’s lead and included a C.A.C. in one of its New York Law issuances in 2003.177 For this bond, the C.A.C. permitted the amendment of payment terms once an eighty five percent consent threshold was obtained. Through the exchange offer, Belize garnered the consent of 87.3% of the holders of this particular bond to amend the payment terms (via C.A.C.s used in conjunction with exit consents) and match them to the terms of the new bonds being issued pursuant to the ex-‐change.178 Thus, the New York bond’s “non-‐complying or non-‐responding credi-‐
173 Id. at 23.
174 See generally Lee C. Buchheit & Elizabeth Karpinski, Belize’s Innovations, 22 BUTTERWORTH’S J. INT’L BANKING & FIN. L. 278 (2007) (discussing Belize’s first exchange offer and how Belize opted to engage bondholders with openness and transparency as a way to show its goodwill towards bond-‐holders).
175 For more information on this tactic see generally Buchheit & Gulati, Exit Consents, supra note 170.
176 Lee C. Buchheit & Elena L. Daly, Minimizing Holdout Creditors: Sticks, in SOVEREIGN DEBT MANAGEMENT, supra note 35, at 15, 18.
177 See Buchheit & Karpinski, supra note 174, at 279 (mentioning how Belize followed Mexico’s lead in 2003).
178 Id. at 279-‐80.
Núm. 4 (2015) COLLECTIVE ACTION CLAUSES FOR PUERTO RICAN BONDS 1227
tors” functionally participated in the exchange and Belize was able to guarantee the restructuring of ninety eight percent of outstanding claims.179
To be sure, other factors, including Belize’s amenable behavior towards creditors, were pivotal to the success of the restructuring. Belize continued to timely pay obligations as they became due, embraced transparency and routinely shared financial information and economic projections with creditors.180 In addi-‐tion, Belize set out a minimum participation requirement (where the offer would not go forward unless a certain number of bondholders accepted the offer) and invited bondholders to form a formal negotiation committee representing fifty one percent of the eligible debt. Even though bondholders failed to meet this requirement, Belize irrespectively opted to negotiate with the committee as a showing of good faith.181 No significant holdout challengers ensued after the offer expired.
Belize’s subsequent 2013 debt exchange offer trod down a similar path.182 The country launched an exchange offer whereby if seventy five percent of the eligi-‐ble debt tendered their old bonds, the C.A.C. would be triggered to modify the remaining untendered bonds.183 After a back and forth negotiation with bond-‐holders resulted in 86.2% of the eligible debt being tendered, Belize subsequent-‐ly activated C.A.C.s to loop in the remaining bondholders, imposed the same terms of the new bonds on the old bonds, and achieved a 100% participation rate.184 This time around, however, the bondholder committee was more engaged and fearlessly used its bargaining leverage to resist Belizean overtures.
When the Belizean government announced that it needed to complete an-‐other restructuring, bondholders quickly responded with skepticism. The extent of debt relief needed by Belize, if any at all, was debatable.185 They were suspi-‐cious of the country’s proffered rationales because economic indicators painted a
179 Duggar, supra note 163, at 36. See MICHELLE ROBINSON, DEBT RESTRUCTURING INITIATIVES PAPER 13 (2010), available at http://michelerobinson.net/yahoo_site_admin/assets/docs/Debt_Restructuring_Initiatives_Paper.330180337.pdf.
180 See, e.g., Buchheit & Karpinski, supra note 174; Information for Creditors, CENT. BANK OF BELIZE, https://www.centralbank.org.bz/financial-‐system/information-‐for-‐creditors (last visited Apr. 16, 2015).
181 See Asonuma et al., supra note 166, at 7-‐10.
182 See Georgia Wells & Katy Burne, Belize Default Looms as Negotiations Continue, WALL ST. JOURNAL (Sept. 19, 2012, 5:24 PM), http://www.wsj.com/articles/SB10000872396390444032404578006440328175304 (discussing the events leading up to Belize’s second debt exchange offer).
183 See Press Release, Cent. Bank of Belize, Belize Launches Debt Exchange Offer (Feb. 15, 2013), available at https://www.centralbank.org.bz/docs/rsh-‐1.7-‐information-‐for-‐creditors/press-‐release-‐-‐-‐belize-‐launches-‐debt-‐exchange-‐offer-‐february-‐15-‐2013.pdf?sfvrsn=2 (describing how C.A.C.s would be used as part of Belize’s 2013 exchange offer).
184 Duggar, supra note 163, at 36.
185 Robin Wigglesworth, Clock Ticks on Belize Debt Restructuring, FIN. TIMES (Nov. 13, 2012, 9:03 PM) [hereinafter Wigglesworth, Clock Ticks], http://www.ft.com/cms/s/0/a0828002-‐2808-‐11e2-‐afd2-‐00144feabdc0.html#axzz3VHc3Nhu0.
1228 REVISTA JURÍDICA UPR Vol. 84
more positive picture than the government portrayed. Put another way, they voiced similar moral hazard concerns invoked by those worried that C.A.C.s cre-‐ate incentives for debtors to impose losses on creditors instead of adopting polit-‐ically unpopular measures. Belize initially extended an offer that entailed losses of up to eighty percent; bondholders were enraged and the negotiations soured.186 Things were further complicated when Belize missed a coupon pay-‐ment and consequently triggered legal remedies for bondholders. The Country was able to secure forbearance of these legal remedies from the creditor commit-‐tee only after paying half of the missed coupon.187 In summary, the tables were turned in favor of bondholders. Notwithstanding their diminished ability to holdout, they nevertheless found in C.A.C.s a medium to stand their ground against an offer they perceived unfair.
Belize’s second restructuring vividly illustrates that many of the concerns voiced by those who view C.A.C.s with suspicion may be overblown.188 C.A.C.s will not do their job unless debtors and bondholders are willing to accept sacri-‐fices. The fact that the creditor committee went as far as challenging Belize’s proposals, forcing the Country to scale back its requests for relief, is a testament to the fact that C.A.C.s do not guarantee that a sovereign has the upper hand during negotiation.189 Neither do C.A.C.s ensure a quick and painless restructur-‐ing.190 In the end, Belize obtained an overall haircut of ten percent, an amount considerably lower than what the Country originally intended from the get-‐go,191 but was yet again spared from significant holdout challenges.
Ultimately, the Belizean experience illustrates that C.A.C.s may actually de-‐liver the ex-‐ante and ex-‐post benefits they purport to provide. Some of the ex-‐ante C.A.C. fears identified as being linked with increased borrowing costs may not necessarily materialize, particularly the possibility of opportunistic behavior by debtors. In both instances, an orderly restructuring was completed in a mat-‐ter of months with little to no resistance from holdouts. Bondholders were able to internalize Belize’s predicament and understand how they benefit by keeping Belize alive as a going concern, even if it meant a loss in the form of debt relief. A
186 Id.; Robin Wigglesworth, Belize Does ‘Superbond’ Deal with Lenders, FIN. TIMES (Feb. 13, 2013, 12:27 PM) [hereinafter Wigglesworth, Belize Does ‘Superbond’], available at http://www.ft.com/intl/cms/s/0/2817d01c-‐75d4-‐11e2-‐b702-‐00144feabdc0.html.
187 Wigglesworth, Clock Ticks, supra note 185.
188 Assuming, of course, that the debtor country will actively avoid an Argentina-‐like conflict with bondholders and values negotiation with creditors to preserve the market’s good will.
189 See Memorandum from the Ministry of Fin. and Econ. Dev., Gov’t of Belize, to Holders of Belize’s U.S. Dollar Step-‐Up Bond Due 2029, Update on the Proposed Restructuring of the Above Captioned Instrument 2-‐3 (Nov. 29, 2012), available at https://www.centralbank.org.bz/docs/rsh-‐1.7-‐information-‐for-‐creditors/memorandum-‐to-‐bondholders-‐29-‐november-‐2012.pdf?sfvrsn=4.
190 See Asonuma et al., supra note 166, at 18 (describing the negotiations between Belize and its bondholders).
191 See Wigglesworth, Belize Does ‘Superbond’, supra note 186 (discussing how this haircut was lower than original expectations).
Núm. 4 (2015) COLLECTIVE ACTION CLAUSES FOR PUERTO RICAN BONDS 1229
valid question remains as to whether Belize secured enough debt relief to avoid a subsequent restructuring, but the presence of C.A.C.s will, at least, allow this conversation to happen, if indeed it needs to occur. Ex-‐post cost-‐reduction bene-‐fits depend on the clause’s interaction with other bond terms and on the active and informed participation of everyone involved in the process. The latter hinges on multiple factors such as the availability of information, how an exchange offer is structured and the ability of bondholders to form a cohesive unit. Because reputational concerns curb the sovereign’s self-‐serving incentives and moral hazards, bondholder recovery expectations need to be commensurate with the distressed reality of the debtor, and C.A.C.s provide the vehicle to shape these expectations. Belize’s experience is therefore evidence that C.A.C.s can check and balance the natural propensity of both parties to pursue their own self-‐interests.
Beyond Belize, states can learn a number of lessons from other sovereign debt restructurings. Most successful restructurings have occurred in the middle of “clear and acute crisis” where it was obvious that debt levels were unsustaina-‐ble.192 Sovereign nations must offer minimally appealing debt exchanges to en-‐courage voluntary participation while still stacking the deck in their favor by threatening default and amending bond terms, among other options.193 Restruc-‐turings also revolve around the composition of instruments and creditors, with special care being taken to minimize local banking exposures.194 In addition, an external official sector entity, such as the I.M.F., has had a role in the restructur-‐ings, suggesting the possibility that the Federal Government might need to get involved if a state seeks this solution.195 Finally, “unpleasant policy adjustments . . . [are] part of any approach to resolving a crisis” since creditors demand policy changes before voluntarily participating in a restructuring.196 Taken together, these lessons spell out the state of the world that should exist in order to effec-‐tively deploy C.A.C.s and reap their benefits.
B. Recent Developments and Proposals for Reform
C.A.C.s still figure prominently in debates on how to improve the efficiency of debt restructurings and minimize creditor coordination costs. Instead of fo-‐cusing on the utility of including C.A.C.s, the modern discourse hones in on al-‐ternative ways to improve their operation. Leading the way are the I.M.F. and the International Capital Markets Association (I.C.M.A.). In September of 2014, the I.M.F. proposed a new model pari passu clause (in response to the Southern Dis-‐
192 Feibelman, supra note 4, at 173 (describing the several conditions that tend to exist at the time a sovereign initiates a debt restructuring).
193 Id. at 174.
194 Id.
195 Id. at 175-‐76.
196 Id. at 176.
1230 REVISTA JURÍDICA UPR Vol. 84
trict of New York’s strict reading of the clause) and C.A.C.197 Relevant to this arti-‐cle is how they propose making C.A.C.s stronger by promoting a more robust aggregation mechanism through single-‐limb voting, enabling a restructuring after the consent of a majority of bondholders across different instruments and eschewing internal quorum requirements within each eligible series.198 This modified C.A.C. would unquestionably make creditor coordination problems easier to overcome by minimizing any holdout’s ability to acquire a blocking position on a given instrument. In doing so, the I.M.F. and I.C.M.A. believe the debt markets benefit from shortening the time period it takes to restructure.199
Naturally, some market actors worry that these reforms will have adverse implications on borrowing costs because they further erode investor protections. Panama, Ghana and El Salvador quickly adopted the pari passu portion of these proposals in recent issuances, yet shied away from the modified C.A.C.; they worried that their pricing effects were yet unknown.200 Perhaps they were trapped by the prisoner’s dilemma —these brand new C.A.C.s were yet untested. Who would be brave enough to test the market’s reaction? Kazakhstan took the risky plunge.
Even after Kazakhstan’s bond issuance was pronounced a blowout success, market commentators were quick to point out that a high demand for Kazakh debt was probably due to the fact that the country hadn’t tapped the debt mar-‐kets in a considerable number of years.201 In other words, the market’s hunger for Kazakhstani debt may have overridden these pricing concerns. It was up to Mex-‐ico, yet again, to take the bold move by including these model clauses in a debt issuance in November of 2014, potentially setting the stage for another shift in how sovereign bond contracts are drafted.202 Given Mexico’s historical reputation
197 See generally INT’L MONETARY FUND, STRENGTHENING THE CONTRACTUAL FRAMEWORK, supra note 165.
198 See id. at 1, 15-‐23 (describing the objectives and operation of the proposed enhanced C.A.C.s).
199 See Landon Thomas Jr., I.M.F. Considers Rules to Force Bondholders to Share Cost of Restruc-‐turing, N.Y. TIMES (Sept. 4, 2014, 6:48 PM), http://dealbook.nytimes.com/2014/09/04/i-‐m-‐f-‐considers-‐rules-‐to-‐force-‐bondholders-‐to-‐share-‐cost-‐of-‐restructuring/?_r=0.
200 Steven Gilmore, Kazakh CAC Adoption Heralds New Era for Sovereign Debt Restructuring, EMERGING MKTS. (Sept. 10, 2014), http://www.emergingmarkets.org/Article/3388638/Kazakh-‐CAC-‐adoption-‐heralds-‐new-‐era-‐for-‐sovereign-‐debt-‐restructuring.html (describing Kazakhstan’s adoption of the new C.A.C.s). See GlobalCapital, Don’t Draw CAC Conclusions from Kazakhstan Blowout, GLOBALCAPITAL (Oct. 7, 2014), http://www.globalcapital.com/article/nfqdtrkbn2s4/don39t-‐draw-‐cac-‐conclusions-‐from-‐kazakhstan-‐blowout (“Though the new documentation would only help borrowers in a restructuring, many issuers have wanted someone else to go first to show that pricing would not be affected, rather than create a two tier market in their own deals.”).
201 See GlobalCapital, supra note 200 (“The demand for this deal says much more about investors’ desire for Kazakhstan paper than their feelings about an appropriate premium for an ICMA Collec-‐tive Action Clause.”).
202 See Landon Thomas Jr., Mexico’s Bold Move on Debt Restructuring Contracts, N.Y. TIMES (Nov. 12, 2014, 12:42 PM), http://dealbook.nytimes.com/2014/11/12/mexicos-‐bold-‐move-‐on-‐debt-‐restructuring-‐contracts/?_r=0.
Núm. 4 (2015) COLLECTIVE ACTION CLAUSES FOR PUERTO RICAN BONDS 1231
as a market leader and trustworthy issuer, it is likely that we will see another evolution in the drafting of boilerplate bond terms.203
From the Commonwealth’s perspective, these reforms illustrate two im-‐portant lessons. First, the Commonwealth could readily use, instead of building from scratch, a C.A.C. carefully structured by a group of experts knowledgeable in the dynamics of creditor collective actions problems. Second, the fact that official sector entities are still pushing for their inclusion, albeit on more debtor friendly terms, provides an argumentative justification on how, as a matter of public policy, the experience with holdouts and coordination problems contin-‐ues to warrant a strong contractual response.204
In the past, the United States Department of the Treasury advocated in favor of using a market-‐oriented approach embodied by C.A.C.s as their preferred method to smoothen restructurings.205 Rejecting calls for an international bank-‐ruptcy framework for sovereigns, the then Secretary of the Treasury, John Snow, recognized that the contractual nature of bond agreements was the root of coor-‐dination problems, but stopped short of backing a government-‐level solution. Instead of allocating authority to an entity outside this contractual relationship, he believed that the parties “must assume responsibility” and find a solution within the four corners of the contract.206 Ergo, his endorsement of the C.A.C. An
203 See Elaine Moore, Mexico’s Move Set to Shake up Bond Market, FIN. TIMES (Nov. 11, 2014, 10:21 PM), http://www.ft.com/intl/cms/s/0/6787610e-‐69b6-‐11e4-‐8f4f-‐00144feabdc0.html#axzz3StS1qW5q (discussing how Mexico could herald a change in how sovereign bond terms are drafted); Daniel Bases, Sovereign Issuers Seek to Thwart Argentina-‐style Attacks by Funds, REUTERS (Mar. 23, 2015) [hereinafter Bases, Sovereign Issuers], http://www.reuters.com/article/2015/03/23/emergingmarkets-‐debt-‐covenants-‐idUSL2N0WM2KB20150323 (quoting market commentators that think that the en-‐hanced C.A.C.s are not having any detrimental effect on an issuer’s credit rating). But see Davide Scigliuzzo, Changes to LatAm Bond Clauses Raise Market Hackles, REUTERS (Mar. 7, 2015), http://in.reuters.com/article/2015/03/06/emergingmarkets-‐bonds-‐cac-‐idINL5N0W804120150306 (discussing how the market may be presently viewing the enhanced C.A.C.s with some reservations).
204 See Bases, Sovereign Issuers, supra note 203 (describing how the new C.A.C.s represent a con-‐cern with the holdout effect on restructurings). See also Joseph E. Stiglitz, How to Keep the Sovereign Bond Vultures at Bay, GULF NEWS (Oct. 15, 2014), http://gulfnews.com/business/analysis/how-‐to-‐keep-‐the-‐sovereign-‐bond-‐vultures-‐at-‐bay-‐1.1398564 (arguing that the fallout from U.S. court rulings makes it difficult for sovereigns to obtain debt relief without unjustly enriching vultures, and war-‐rants a better contractual framework to counter holdouts); Anna Gelpern, A Sensible Step to Mitigate Sovereign Bond Dysfunction, PETERSON INST. FOR INT’L ECON. (Aug. 29, 2014, 4:57 PM), http://blogs.piie.com/realtime/?p=4485 (arguing that the dysfunctional nature of sovereign debt justifies strengthening C.A.C.s).
205 See Hagan, supra note 28, at 390-‐94 (discussing the U.S. Treasury’s opposition to the mecha-‐nism proposed by the I.M.F. and its endorsement of C.A.C.s as their preferred mechanism).
206 Statement, John W. Snow, Sec’y, U.S. Dept. of the Treasury, Statement at the Meeting of the International Monetary and Financial Committee (IMFC) 2 (Apr. 12, 2003), available at https://www.imf.org/external/spring/2003/imfc/state/eng/usa.htm (last visited Apr. 18, 2015). The Secretary stated that:
There can at times be collective action problems that prevent a prompt, orderly resolution of a sovereign debt crisis. The source of these problems lies in the relationships and agreements of debtors and their creditors. It is these parties, not an international organiza-‐tion, that must assume responsibility for the solution.
1232 REVISTA JURÍDICA UPR Vol. 84
obvious question is whether such a stance will be revived in the context of a state debt crisis. Perhaps a C.A.C.’s ability to address lingering federalism concerns will do the trick.207
V. C .A.C.S IN PUERTO RICAN BONDS: IMPLICATIONS FOR THE FUTURE
Debt management has an indelible moral component; borrowing money from the public debt markets implicates concerns beyond short and medium term financing.208 The creditworthiness of an entire country or state, and by vir-‐tue of their functional roles, their citizens, will be at stake. Unexpected fiscal developments or the baggage of sustained years of fiscal mismanagement can be a tough pill to swallow as initial attempts to quell fiscal crises shift most of the burden of recovery on to ordinary citizens. Without a doubt, it is difficult to establish the proper extent or scope of remedial fiscal measures a government must adopt before it can confidently stand before the court of creditor opinion and ask bondholders to absorb losses. As it stands, the structure of debt financ-‐ing creates perverse incentives: a debtor receives quick and easy financial gratifi-‐cation at the expense of those who, in the far-‐flung future, have to stomach “the distasteful residuum of . . . debt—the need to pay it back.”209 The extent to which past Commonwealth governments irresponsibly over-‐borrowed is for historians to mull upon. The task right now is to figure out how to confine its effects in the future. Equipping these debt instruments with the proper contractual tools for modifying them if they become prohibitively unsustainable is one way to do so.
This article reaffirms what multiple scholars have argued before: one such tool is the C.A.C., a contractual mechanism well known in the sovereign debt context but relatively untested in the municipal debt markets. Because the American municipal debt market has yet to embrace C.A.C.s, it is understanda-‐ble why policymakers would be naturally reluctant to include them in future issuances. They might fear that doing so will be tantamount to hoisting a red flag, setting off fear and trepidation among potential investors that forecloses market access.
Surely, no reasonable issuer wants to signal a willingness to impose losses on the bondholders they rely upon to finance essential public services. Being the first American municipal issuer to embrace C.A.C.s would likely seize the mar-‐ket’s attention. Some may question the wisdom of this policy, wondering wheth-‐
Id.
207 A market-‐oriented approach espoused by C.A.C.s may stay the hand of federal involvement and allow states to solve fiscal crises on their own terms. In addition, the Federal Government will likely not want to set awkward precedent and bail out a state.
208 See Lee C. Buchheit, Sovereign Debt in the Light of Eternity, in SOVEREIGN DEBT MANAGEMENT, supra note 35, at 463 (discussing how the legacy of debt incurrence should prompt decision makers to evaluate borrowing through a moral, legal, financial and political lens beyond immediate financial benefits.).
209 Id. at 464.
Núm. 4 (2015) COLLECTIVE ACTION CLAUSES FOR PUERTO RICAN BONDS 1233
er it could unravel a relatively stable municipal debt market by introducing an element of uncertainty. But an argument could be made that there is already enough uncertainty when states and Commonwealth issuers have no formal bankruptcy mechanism. In other words, the fact that Puerto Rico cannot rely on a formal bankruptcy regime should greatly minimize these signaling concerns. The inclusion of C.A.C.s can be justified as nothing more than an ex-‐ante at-‐tempt to avoid the greater evil of a messy default, where the lack of a clear framework can unleash negative spillovers that harm debtors and creditors alike.
An issuer’s desire to free itself from this prisoner’s dilemma may be aug-‐mented by an empirical and practical understanding of how C.A.C.s address the-‐se spillovers, an understanding informed by the experience of the sovereign debt markets. It is fair to assume that how C.A.C.s concurrently protect and constrain the interests of both debtors and creditors is now, more than ever, better under-‐stood as an important stepping-‐stone in the drafting of more efficient bond terms that respond to the practical realities of state and municipal debt financ-‐ing. Given the fiscal problems currently faced by many American jurisdictions, this shift is a long overdue Sputnik moment.
Creditor coordination problems have become an important element factored into the calibration of risk by bondholders and sovereign debtors. All the empiri-‐cal studies discussed in this article suggest that investors universally recognize that C.A.C.s make it easier for a debtor to obtain debt relief. They are not entire-‐ly in agreement, however, on the extent that cost of capital is sensitive to the risk of downward adjustment they represent.210 A decision to move Puerto Rican bonds closer to the sovereign C.A.C.-‐model will undeniably invite a recalibration of risk.211 Bonds that lack C.A.C.s are assumed to require unanimous consent before amending payments terms, whereas C.A.C.s make it easier to impose haircuts on bondholders even though they do not agree. But risk is in the eye of the beholder, and yield spreads may simply reflect the market’s perceptions of the Commonwealth’s economic fundamentals at a given point in time. Because bondholders, as a whole, have idiosyncratic interest and motivations, some of them might not even think twice about whether a C.A.C. represents a higher risk of default. Nevertheless, investors who value a higher likelihood of being repaid in full, or to their ability to resist a restructuring, appear to be willing to pay a holdout premium in exchange for a safer investment.212
As explained in Part II.B., a holdout premium essentially means that borrow-‐ing costs are lowered whenever a restructuring is perceived to be untenable. For example, a holdout premium can exist when there are two comparable, yet dis-‐tinct, debt instruments from the same issuer that are nonetheless subject to dif-‐ferent governing laws. But there is some evidence that a holdout premium may
210 Stephen J. Choi & Mitu Gulati, From Pigs to Hogs 2 (Jan. 7, 2015) (draft), available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2434272 (“The existing literature on sovereign bonds is unclear on the answer to the question of whether contract terms are priced.”).
211 See supra notes 39-‐40 and accompanying text.
212 See generally discussion in Choi et al., Pricing Terms, supra note 30.
1234 REVISTA JURÍDICA UPR Vol. 84
be more of a function of situations, like the one faced by Puerto Rico, of “severe debt distress.”213 Given the lack of contractual framework to facilitate the ad-‐justment of burdensome obligations, should Puerto Rico nevertheless embrace C.A.C.s and forgo a holdout premium? Regardless of whether Puerto Rican bonds currently enjoy a holdout premium vis-‐à-‐vis the governing law of particu-‐lar issuance, where a Puerto Rico bond issued under the laws of another jurisdic-‐tion would be presumably cheaper to issue, the debate fundamentally boils down to whether Puerto Rico should continue being vulnerable to holdout credi-‐tors.
The moral element of this debate need not be repeated, otherwise this arti-‐cle would belabor the obvious, but the practical component does force a reckon-‐ing with reality. Such a debate needs to be informed by several constraints, in-‐cluding the need to ensure continuous market access and the stark reality that fiscal crises may yet again haunt Puerto Rico in the future. “It is fatuous to trust that a state of perpetual benignity will reign in the realms of politics, finance and the natural world.”214 Absent a formal bankruptcy alternative, collective action problems may be an obstacle for Puerto Rico to effectively tackle future fiscal crises. All things being equal, the contractual framework espoused by C.A.C.s should be preferred over perennial uncertainty on how a cash-‐strapped Com-‐monwealth could secure relief.
We should never divorce recent historical developments from our under-‐standing of how C.A.C.s may influence borrowing costs. The Argentine holdout litigation, Greece’s debt crisis and Belize’s successful restructurings are just a few examples of episodes that energized the utility of C.A.C.s. However, not every commentator believes that C.A.C.s are as needed or effective as advocates make them out to be.215 Belize certainly proves them right, to a point. Belize did not obtain the debt relief it originally hoped for, but was spared from costly fallout.216 To be sure, it is entirely possible that Belize might hypothetically need another round of restructuring in the near future because it was not able to convince
213 Marcos Chamon et al., Foreign Law Bonds: Can They Reduce Borrowing Costs? 3 (Oct. 7, 2014) (draft), available at https://editorialexpress.com/cgi-‐bin/conference/download.cgi?db_name= RES2015&paper_id=901 (“Our main result is that a foreign-‐law premium exists, but it only becomes significant and sizeable in periods of severe debt distress, with a likely debt restructuring on the horizon.”).
214 Lee C. Buchheit, Cleary Gottlieb Steen & Hamilton LLP, Sovereign Fragility, Keynote Address at the Adam Smith Business School at the University of Glasgow’s Emergent Paradigms in Sovereign Debt: Law, Economics and Finance Workshop 2-‐3 (May 23, 2014), draft available at http://www.gla.ac.uk/media/media_332913_en.pdf.
215 See John A. E. Pottow, Mitigating the Problem of Vulture Holdout: International Certification Boards for Sovereign-‐Debt Restructurings, 49 TEX. INT’L L.J. 221, 224 (2014) (questioning the ability of C.A.C.s to effectively counter the problem of holdout creditors). See also Ran Bi et al., The Problem that Wasn’t: Coordination Failures in Sovereign Debt Restructurings 19-‐20 (I.M.F., Working Paper No. 11/265, 2011), available at https://www.imf.org/external/pubs/ft/wp/2011/wp11265.pdf (concluding that the bondholder coordination problem may not be as problematic in practice as many believe).
216 See discussion in Part III.A. on Belize’s second restructuring and the challenges Belize encoun-‐tered from bondholder resistance notwithstanding its ability to use a C.A.C.
Núm. 4 (2015) COLLECTIVE ACTION CLAUSES FOR PUERTO RICAN BONDS 1235
bondholders to internalize the requisite losses in earlier rounds of negotiation. But the fact that its outstanding bonds have C.A.C.s at least allows Belize to con-‐sider reengaging creditors for relief without balking at the prospects of unneces-‐sarily hefty transaction costs. It is ultimately a question of balancing competing interests. Commonwealth policymakers should consider increased borrowing costs stemming from moral hazard concerns in light of the ex-‐ante benefits of an orderly mechanism.
On the one hand, increased costs of capital associated with moral hazard may make it expensive for the Commonwealth to finance much-‐needed essential public services. There is something to be said, however, about C.A.C.s staying the hand of reckless borrowing that leads to debt overhang and skittish uncer-‐tainty. A frantic issuer could otherwise be tempted to engage in short-‐term bor-‐rowing at usurious prices to delay much needed structural reforms or postpone a default. In the long run, risk premiums associated with C.A.C.s can therefore induce better borrowing practices when a mechanism erodes incentives to pro-‐long a fiscal crisis.
On the other hand, the livelihoods of future generations of Puerto Ricans will be fundamentally tied to the decisions made by past policymakers. Including C.A.C.s in future issuances can amount to an implicit recognition that present fortunes will not necessarily extend to the future. It is hard to determine wheth-‐er ten or twenty years from now the Commonwealth will be able to service debt obligations while fulfilling its public mandate to provide for the health, welfare and safety of Puerto Ricans. C.A.C.s have the unique benefit that they can be reused to respond to a pliable market. Solvency and liquidity, necessary to safe-‐guard an ability to honor debt obligations, are concepts tied to the vagaries of economic growth and an unpredictable capacity to generate sufficient revenues to ensure a positive primary balance. If used responsively, C.A.C.s permit ad-‐justments in a market-‐oriented manner that recognizes the inherent unpredict-‐ability of economic prospects.
All things being equal, Commonwealth policymakers should also favor the inclusion of C.A.C.s in future issuances as an ex-‐post tool to facilitate coordina-‐tion, communication and negotiation with bondholders. But they should not be easily enticed unless they understand that the benefits of C.A.C.s depend on many factors. Namely, a willingness to let others monitor and check their behav-‐ior before and during a restructuring. Empirical studies have found a link be-‐tween the extent of debt relief an issuer requests and the subsequent pervasive-‐ness of the bondholder coordination problem.217 It might therefore be rational to include C.A.C.s in future issuances, a flexible shield against these problems, in order to allow the Commonwealth to deal with future scenarios where the need-‐ed debt relief and potential coordination obstacles are impossible to anticipate.
It is also important to note that how a particular bondholder views C.A.C.s, as either a tool or a threat, is intrinsically tied to his or her investment agenda.
217 See Ran Bi et al., supra note 215, at 19 (concluding that the bondholder coordination is tied, in part, to the extent of debt relief an issuer is requesting).
1236 REVISTA JURÍDICA UPR Vol. 84
Compulsive holdouts will continue to see these clauses as the bane of their exist-‐ence. Many of them are willing to stall a restructuring in order to extract favora-‐ble payment for themselves, sometimes to the detriment of fellow bondholders. In contrast, other investors may see the advantage of better coordination and relative procedural expediency in a positive light. As long as the Commonwealth refrains from opportunistic behavior or from proposing unfair adjustments to their obligations, C.A.C.s will be a useful mechanism to de-‐fang, so to speak, holdout investors, while forcing Puerto Rico to engage bondholders in a concilia-‐tory, pragmatic manner under the cover of a somewhat democratic framework.
The majority of the borrowing cost studies implicitly or explicitly point to the creditworthiness of a sovereign issuer as an important predictor of an in-‐creased cost of capital associated with C.A.C.s.218 As long as Puerto Rican debt continues to be rated below investment grade, the inclusion of C.A.C.s may im-‐pact borrowing costs to some extent. This increase will presumably be due to heightened concerns about Puerto Rico’s incentives to shift losses unto creditors at the sign of any fiscal impasse. C.A.C.s can perhaps be understood as the func-‐tional equivalent of taking out an insurance policy against fiscal distress. The higher the likelihood of default, the higher the price of this policy appears to be. However, the underlying economic fundamentals of a debtor may already exert strong influence on borrowing prices independent on whether bonds include C.A.C.s. The speculative nature of projecting a hypothetical scenario of future distress makes measuring the benefits of this insurance elusive. But the Belizean experience discussed in Part III.A. lends credence to the idea that moral hazard, free riding and holdout problems can be controlled in manner that nonetheless preserves a bondholder’s bargaining leverage. The caveat being that bondholders need to be able to form a cohesive and informed unit and a debtor must be committed to prefer negotiation over confrontation, and perhaps be willing to accept less than optimal debt relief, as the means to avert an inefficient restruc-‐turing or costly default.
Traditional assumptions on how consent thresholds dilute or enhance the debtor’s leverage during debt negotiations may not necessarily apply in the Commonwealth’s current circumstances.219 When bondholder identities and agendas are heterogeneous, a higher consent threshold obviously weakens the debtor’s ability to obtain advantageous debt relief terms because the affirmative consent of a greater number of creditors needs to be secured. As a practical, if not logical, matter, the Commonwealth should prefer lower consent thresholds when bondholders are dispersed, bonds widely held, and thus make it harder to organize creditors. In contrast, a higher consent threshold could be beneficial to prospective investors in situations where bondholders tend to have large con-‐centrated holdings. Recent evidence points to a shift in Commonwealth bond-‐holders from traditional mom and pop retail investors to specialized hedge
218 See discussions in Part II accompanying Eichengreen & Mody, Would Collective Action Clauses, supra note 114; Choi et al., Pricing Terms, supra note 30; Bardozzetti & Dottori, supra note 149.
219 See, e.g., Billington, supra note 35; Bradley & Gulati, supra note 41.
Núm. 4 (2015) COLLECTIVE ACTION CLAUSES FOR PUERTO RICAN BONDS 1237
funds.220 Assuming Puerto Rico needs to rely on these distinct bondholders for financing, at least in the short term, a higher consent threshold would probably calm their concerns about losing bargaining power. Giving them a stronger hand in any future restructuring negotiations would likely minimize how they price the risk of abuse.
Beyond these threshold considerations, perhaps a defensible argument can be made that the current pool of potential bondholders has nothing to fear from these clauses in the short-‐term. C.A.C.s help game the uncertainty of liquid bondholder identities by accounting for the open market trading of these bonds amongst the anonymous buyers of the secondary market. As long as prospective bondholders continue to be like-‐minded institutional investors who strategically amass substantial holdings looking for high yield returns, the dilution of their bargaining leverage under C.A.C.s is likely negligible. Assuming, of course, that these institutional investors can effectively coalesce under the same banner. Di-‐lution fears are therefore more realistic when Puerto Rican bonds are held by bondholders who are relatively dispersed and heterogeneous, therefore less prone to identify, and act, as one fraternal agglomeration of creditors.
Nevertheless, it is undeniable that bondholders naturally prefer to be paid back every cent. But to the extent that a debtor’s fiscal crisis is particularly acute, or debt service obligations become blatantly unsustainable, a collective benefit is derived from preserving, rather than limiting, the debtor’s ability to pay. C.A.C.s should therefore be given greater weight when viewed in light of their ability to avoid a default. As a general matter, defaults have multiple macroeconomic con-‐sequences that can cripple the debtor’s economy. A defaulting issuer will be shut out of the capital markets,221 disabling a critical source of capital to finance day-‐to-‐day expenditures, ranging from payroll to welfare benefits. The litigation that ensues tarnishes the debtor’s reputation and would push the hands of courts being asked to provide relief to bondholders. Spillovers could also affect the debtor’s domestic financial markets, straining the local banking system and sty-‐mying output, trade and investment.222 To be sure, C.A.C.s do not make a default impossible. Bondholders are still able to actively resist an issuer’s proposals and block relief by rejecting a proposed modification. They do, however, provide incentives that thrust both sides to a negotiation table where information can be exchanged and both sides can signal to each other their preferences and con-‐cerns. In light of the Commonwealth’s current predicament, where the contrac-‐tual framework doesn’t lend itself for efficient negotiation and there is no formal legal mechanism to fall back upon, any alternative is preferable to nothing.
220 See Weidemaier & Gulati, Sovereign Debt, supra note 32.
221 This idea is not without its share of detractors. See Wright, supra note 79, at 158; STURZENEGGER & ZETTELMEYER, supra note 22, at 35.
222 See Wright, supra note 79, at 159; Das et al., supra note 168, at 66 (summarizing the economic costs and effects of a restructuring or default); STURZENEGGER & ZETTELMEYER, supra note 22, at 49-‐52 (discussing empirical evidence on the costs of default).
1238 REVISTA JURÍDICA UPR Vol. 84
Notwithstanding the arguments of this article, C.A.C.s should not be thought of as a fixer standing ready with alacrity. What these clauses will give Puerto Rico with one hand, they will continue to withhold from the other. While C.A.C.s minimize the threat of holdouts and facilitate a more cost-‐efficient re-‐structuring, they do not ensure a quick, painless and easy fix to a liquidity or solvency problem. Holdouts will still be able to marshal enough power to block a restructuring, albeit at a higher cost to them. The utility of these clauses will be maximized only to the extent that they are widely included in future issuances and, some would argue, incorporate an aggregation mechanism to loop in multi-‐ple issuances into one restructuring. Until the entire debt stock of non-‐C.A.C.s bonds is retired or matures, collective action problems will linger as vestiges of a time when the drafters of Puerto Rican bonds likely did not foresee the possibil-‐ity that Puerto Rico might need debt relief. Very few imagined the possibility of a buoyant economy enduring the crisis we now face. Beyond Puerto Rico, the pro-‐spect of an American state defaulting, with no way to avoid the chaos that could ensue, is alive and urgent. Puerto Rico’s government may well become a labora-‐tory for the design of legal and contractual solutions to the fiscal problems grip-‐ping other American jurisdictions.223 The scope, extent and effects of the current debt crisis should motivate policymakers to think about ways to provide Puerto Rico with the means to address future fiscal crises in a sensible manner.224 Going forward, C.A.C.s should be on the table.
223 See David Skeel, A Puerto Rican Solution for Illinois, WALL ST. JOURNAL (Aug. 3, 2014, 4:51 PM), http://online.wsj.com/articles/david-‐skeel-‐a-‐puerto-‐rican-‐solution-‐for-‐illinois-‐1407099069 (discuss-‐ing how Puerto Rico may become a trailblazer, adopting strategies that could be emulated by states with similar fiscal problems, like California and Illinois).
224 Gelpern, supra note 204 (paraphrasing Professor Gelpern’s choice of words).